commit 3d83ba0408d669b46a0f357db708796f51611e97 Author: ModelHub XC Date: Tue Jun 2 03:08:16 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: langtech-innovation/salamandra-7b-instruct-tools-20260108 Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..ac7b0fe --- /dev/null +++ b/.gitattributes @@ -0,0 +1,41 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +model-00002-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00003-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00004-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00001-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text +salamandra-7b-tools-16k-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..4349852 --- /dev/null +++ b/README.md @@ -0,0 +1,148 @@ +--- +license: apache-2.0 +library_name: transformers +pipeline_tag: text-generation +--- + +> [!WARNING] +> **WARNING:** This is a language model that has undergone instruction tuning for conversational settings that exploit function calling capabilities. It has not been aligned with human preferences. As a result, it may generate outputs that are inappropriate, misleading, biased, or unsafe. These risks can be mitigated through additional post-training stages, which is strongly recommended before deployment in any production system, especially for high-stakes applications. + +### How to use +``` +from datetime import datetime +from transformers import AutoTokenizer, AutoModelForCausalLM +import transformers +import torch + +model_id = "BSC-LT/salamandra-7b-instruct" + +text = "What is the weather like in Paris today?" + +tokenizer = AutoTokenizer.from_pretrained(model_id) +model = AutoModelForCausalLM.from_pretrained( + model_id, + device_map="auto", + torch_dtype=torch.bfloat16 + ) + +message = [ { "role": "user", "content": text } ] + +tools = [{ + "type": "function", + "name": "get_weather", + "description": "Get current temperature for a given location.", + "parameters": { + "type": "object", + "properties": { + "location": { + "type": "string", + "description": "City and country e.g. Bogotá, Colombia" + } + }, + "required": [ + "location" + ], + "additionalProperties": False + } +}] + + +prompt = tokenizer.apply_chat_template( + message, + tokenize=False, + add_generation_prompt=True, + tools=tools +) + +inputs = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt") +outputs = model.generate(input_ids=inputs.to(model.device), max_new_tokens=1000) + +print(tokenizer.decode(outputs[0], skip_special_tokens=True)) +``` + +#### Output: +```text + +{"name": "get_weather", "arguments": {"location": "Paris, France"}} + +``` + + +### Deploy with vllm +**Deploy the model using vllm docker image.** +``` +docker run --runtime nvidia --gpus all \ + -v ~/.cache/huggingface:/root/.cache/huggingface \ + --env "HUGGING_FACE_HUB_TOKEN=" \ + -p 80:80 \ + vllm/vllm-openai:latest \ + --model BSC-LT/salamandra-7b-instruct-tools \ + --enable-auto-tool-choice \ + --tool-call-parser hermes \ + --max_model_len 8196 \ + --port 80 +``` + +**Then use it with openai api** +``` +pip install openai +``` +``` +from openai import OpenAI +client = OpenAI( + base_url="http://localhost:8080/v1/", + api_key="hf_xxxx" + ) + +models = client.models.list() +model = models.data[0].id + +system_message = "" +messages = [{ "role": "system", "content": system_message}] if system_message else [] +messages.append( {"role":"user", "content": "What is the weather like in Paris today?"}) +print(messages) +chat_completion = client.chat.completions.create( + model=model, + tools=tools, + messages=messages, + stream=False, + max_tokens=1000, + temperature=0.1, + frequency_penalty=0.2, +) + +msg = chat_completion.choices[0].message + +# --- HANDLE TOOL CALL OR NORMAL CONTENT --- + +if not getattr(msg, "tool_calls", None): + # Normal assistant message + print(msg.content) + + messages.append({ + "role": "assistant", + "content": msg.content + }) + +else: + # Assistant tool call message + print(msg.tool_calls) + + messages.append({"role": "assistant", "tool_calls": msg.tool_calls}) + + # --- Fake tool execution example --- + tool_call = msg.tool_calls[0] + # Example: handle the get_weather tool + if tool_call.function.name == "get_weather": + # Fake tool result (this would come from your actual backend) + fake_tool_result = '{"temperature": 18, "unit": "C", "description": "Partly cloudy in Paris"}' + + # Append the tool result message so the model can use it in the next turn + messages.append({ + "role": "tool", + "tool_call_id": tool_call.id, + "name": tool_call.function.name, + "content": fake_tool_result, + }) +``` + diff --git a/added_tokens.json b/added_tokens.json new file mode 100644 index 0000000..d80ba36 --- /dev/null +++ b/added_tokens.json @@ -0,0 +1,8 @@ +{ + "": 256003, + "": 256000, + "": 256002, + "": 256004, + "": 256001, + "": 256005 +} diff --git a/config.json b/config.json new file mode 100644 index 0000000..f4d69d1 --- /dev/null +++ b/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "/gpfs/projects/bsc88/hf-models/Salamandra-7b_pre-1.3-160k_sft-2.0_openlicenses_ankush", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 1, + "eos_token_id": 2, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 11008, + "max_position_embeddings": 16384, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 20.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 10000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.45.0", + "use_cache": true, + "vocab_size": 256006 +} diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..3151261 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,8 @@ +{ + "_from_model_config": true, + "bos_token_id": 1, + "eos_token_id": 5, + "max_length": 8192, + "pad_token_id": 0, + "transformers_version": "4.45.0" +} diff --git a/model-00001-of-00004.safetensors b/model-00001-of-00004.safetensors new file mode 100644 index 0000000..3476670 --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac37ad2828e23f3079fcc6b0486381dae5b1ff6b421ef78b8c7b8529613a9e94 +size 4983022200 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..dfe6e4c --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15ef894def74ba0e39bd990ea0f27b31f6cfdde643cb3bdbd4fae49f26e980c9 +size 4995660232 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..75f46b4 --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:468384d9b1708d3095dbcb67790d81b30e667ca061b06fe3923a05b06af1bd5d +size 3460482936 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..5e49777 --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f39a3c8cd898bd6b13d4f2e117163c1b4c5c15e3f89ba86c50b42a2a4c2885e3 +size 2097201280 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000..024dd8d --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,298 @@ +{ + "metadata": { + "total_size": 15536332800 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.norm.weight": "model-00003-of-00004.safetensors" + } +} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..c909875 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,28 @@ +{ + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>" + ], + "bos_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": "<|im_end|>", + "pad_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "unk_token": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..961ba88 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:088c3fef1823cc93b6d06d1ae0351ffd24af034d12a86081ea725b6d2b4f510e +size 37008085 diff --git a/tokenizer.model b/tokenizer.model new file mode 100644 index 0000000..3307f8d --- /dev/null +++ b/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fa490e57cebce5cb1a0a5b1a5d3fa4de05aee53dc3a44791f1c3401db44d802d +size 4813274 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..cf7574c --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,1152 @@ +{ + "add_bos_token": false, + "add_eos_token": false, + "add_prefix_space": false, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "3": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "4": { + "content": "<|im_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "5": { + "content": "<|im_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "6": { + "content": "<|reserved_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "7": { + "content": "<|reserved_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "8": { + "content": "<|reserved_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "9": { + "content": "<|reserved_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "10": { + "content": "<|reserved_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "11": { + "content": "<|reserved_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "12": { + "content": "<|reserved_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "13": { + "content": "<|reserved_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "14": { + "content": "<|reserved_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "15": { + "content": "<|reserved_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "16": { + "content": "<|reserved_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "17": { + "content": "<|reserved_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "18": { + "content": "<|reserved_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "19": { + "content": "<|reserved_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "20": { + "content": "<|reserved_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "21": { + "content": "<|reserved_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "22": { + "content": "<|reserved_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "23": { + "content": "<|reserved_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "24": { + "content": "<|reserved_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "25": { + "content": "<|reserved_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "26": { + "content": "<|reserved_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "27": { + "content": "<|reserved_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "28": { + "content": "<|reserved_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "29": { + "content": "<|reserved_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "30": { + "content": "<|reserved_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "31": { + "content": "<|reserved_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "32": { + "content": "<|reserved_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "33": { + "content": "<|reserved_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "34": { + "content": "<|reserved_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "35": { + "content": "<|reserved_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "36": { + "content": "<|reserved_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "37": { + "content": "<|reserved_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "38": { + "content": "<|reserved_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "39": { + "content": "<|reserved_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "40": { + "content": "<|reserved_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "41": { + "content": "<|reserved_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "42": { + "content": "<|reserved_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "43": { + "content": "<|reserved_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "44": { + "content": "<|reserved_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "45": { + "content": "<|reserved_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "46": { + "content": "<|reserved_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "47": { + "content": "<|reserved_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "48": { + "content": "<|reserved_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "49": { + "content": "<|reserved_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "50": { + "content": "<|reserved_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "51": { + "content": "<|reserved_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "52": { + "content": "<|reserved_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "53": { + "content": "<|reserved_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "54": { + "content": "<|reserved_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "55": { + "content": "<|reserved_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "56": { + "content": "<|reserved_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "57": { + "content": "<|reserved_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "58": { + "content": "<|reserved_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "59": { + "content": "<|reserved_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "60": { + "content": "<|reserved_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "61": { + "content": "<|reserved_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "62": { + "content": "<|reserved_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "63": { + "content": "<|reserved_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "64": { + "content": "<|reserved_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "65": { + "content": "<|reserved_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "66": { + "content": "<|reserved_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "67": { + "content": "<|reserved_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "68": { + "content": "<|reserved_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "69": { + "content": "<|reserved_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "70": { + "content": "<|reserved_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "71": { + "content": "<|reserved_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "72": { + "content": "<|reserved_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "73": { + "content": "<|reserved_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "74": { + "content": "<|reserved_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "75": { + "content": "<|reserved_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "76": { + "content": "<|reserved_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "77": { + "content": "<|reserved_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "78": { + "content": "<|reserved_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "79": { + "content": "<|reserved_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "80": { + "content": "<|reserved_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "81": { + "content": "<|reserved_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "82": { + "content": "<|reserved_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "83": { + "content": "<|reserved_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "84": { + "content": "<|reserved_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "85": { + "content": "<|reserved_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "86": { + "content": "<|reserved_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "87": { + "content": "<|reserved_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "88": { + "content": "<|reserved_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "89": { + "content": "<|reserved_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "90": { + "content": "<|reserved_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "91": { + "content": "<|reserved_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "92": { + "content": "<|reserved_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "93": { + "content": "<|reserved_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "94": { + "content": "<|reserved_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "95": { + "content": "<|reserved_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "96": { + "content": "<|reserved_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "97": { + "content": "<|reserved_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "98": { + "content": "<|reserved_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "99": { + "content": "<|reserved_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "100": { + "content": "<|reserved_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "101": { + "content": "<|reserved_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "102": { + "content": "<|reserved_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "103": { + "content": "<|reserved_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "104": { + "content": "\\r", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "105": { + "content": "▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "106": { + "content": "▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "107": { + "content": "▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "108": { + "content": "▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "109": { + "content": "▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "110": { + "content": "▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "111": { + "content": "▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "112": { + "content": "▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "113": { + "content": "▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "114": { + "content": "▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "115": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "116": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "117": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "118": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "119": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "120": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "121": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "122": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "123": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "124": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "125": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "126": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "127": { + "content": "▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "128": { + "content": "\t\t", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "129": { + "content": "\t\t\t", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "130": { + "content": "\t\t\t\t", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "131": { + "content": "\t\t\t\t\t", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "132": { + "content": "\t\t\t\t\t\t", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "133": { + "content": "\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "134": { + "content": "\n\n\n", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "256000": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": false + }, + "256001": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": false + }, + "256002": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": false + }, + "256003": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": false + }, + "256004": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": false + }, + "256005": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": false + } + }, + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>" + ], + "bos_token": "", + "chat_template": "{%- set tools = tools if tools is defined else None -%}\n{%- set date_string = date_string if date_string is defined else \"1 Sep 2024\" -%}\n\n{%- set system_message = messages[0].content if messages[0].role == \"system\" else \"\" -%}\n{%- if messages[0].role == \"system\" -%}\n {%- set messages = messages[1:] -%}\n{%- endif -%}\n\n{%- if not tool_prompt -%}\n {%- set tool_prompt = \"For each function call return a json object with function name and arguments within tags with the following schema:\n\n{\\\"name\\\": , \\\"arguments\\\": }\n\" -%}\n{%- endif -%}\n\n{%- if system_message or tools -%}\n {{- '<|im_start|>system\n'}}\n{%- endif -%}\n\n{%- if system_message %}\n {{- system_message + \"\n\"}}\n{%- endif -%}\n\n{%- if tools -%}\n {{- \"You have function-calling capabilities. You are provided with function signatures within XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions.\n\" }}\n {{- \"\n\" }}\n {{- tools }}\n {{- \"\n\n\" }}\n {{- tool_prompt -}}\n{%- endif -%}\n\n{%- if system_message or tools -%}\n {{- '<|im_end|>\n'}}\n{%- endif -%}\n\n{# Main message loop #}\n{%- for message in messages -%}\n {%- if message.role == \"user\" or message.role == \"assistant\" or message.role == \"tool\" -%}\n {%- if loop.first and message.role != \"user\" -%}\n {{ raise_exception(\"Invalid sequence: The first message role must be 'user' after 'system' if provided .\") }}\n {%- endif -%}\n\n {%- if not loop.first and message.role in [\"user\", \"assistant\"] and message.role == loop.previtem.role -%}\n {{ raise_exception(\"Invalid sequence: Consecutive messages cannot have the same role ('user' or 'assistant').\") }}\n {%- endif -%}\n\n {%- if message.role == \"user\" and not loop.first and loop.previtem.role != \"assistant\" -%}\n {{ raise_exception(\"Invalid sequence: A 'user' message must be preceded by an 'assistant' message.\") }}\n {%- endif -%}\n\n {%- if message.role == \"tool\" and not loop.first and loop.previtem.role not in [\"assistant\", \"tool\"] -%}\n {{ raise_exception(\"Invalid sequence: A 'tool' message must be preceded by 'assistant' or 'tool'.\") }}\n {%- endif -%}\n {%- else -%}\n {{- raise_exception(\"Invalid role detected: only 'user', 'assistant', or 'tool' roles are accepted.\") }}\n {%- endif -%}\n {%- if message.role == \"user\" or (message.role == \"assistant\" and message.tool_calls is not defined) -%}\n {{- '<|im_start|>' + message.role + '\n' + message.content | trim + '<|im_end|>\n'}}\n {%- elif message.role == \"assistant\" -%}\n {{- '<|im_start|>' + message.role }}\n {%- for tool_call in message.tool_calls -%}\n {{ '\n\n' }}\n {%- if tool_call.function -%}\n {\"name\": \"{{ tool_call.function.name }}\", \"arguments\": {{ tool_call.function.arguments | tojson }} }\n {%- else -%}\n {\"name\": \"{{ tool_call.name }}\", \"arguments\": {{ tool_call.arguments | tojson }} }\n {%- endif -%}\n {{ '\n' }}\n {%- endfor -%}\n {{- '<|im_end|>\n' }}\n {%- elif message.role == \"tool\" -%}\n {%- if loop.previtem and loop.previtem.role != \"tool\" -%}\n {{- '<|im_start|>tool\n' }}\n {%- endif -%}\n {{- '\n' }} \n {{- message.content }}\n {{- '\n\n' }}\n {%- if loop.last or loop.nextitem.role != \"tool\" -%}\n {{- '<|im_end|>\n'}}\n {%- endif -%}\n {%- endif -%}\n{%- endfor -%}\n\n{# Prompt for assistant generation if needed #}\n{%- if add_generation_prompt -%}\n {{- '<|im_start|>assistant\n' }}\n{%- endif -%}", + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "legacy": true, + "model_max_length": 8192, + "pad_token": "", + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "unk_token": "", + "use_default_system_prompt": false +} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..4d562bb --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,29827 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0, + "eval_steps": 100, + "global_step": 4207, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0002376990729736154, + "grad_norm": 6.243001215724618, + "learning_rate": 7.874015748031497e-08, + "loss": 1.0469, + "step": 1 + }, + { + "epoch": 0.0004753981459472308, + "grad_norm": 50.90334593084712, + "learning_rate": 1.5748031496062994e-07, + "loss": 1.8061, + "step": 2 + }, + { + "epoch": 0.0007130972189208462, + "grad_norm": 141.34422809531102, + "learning_rate": 2.362204724409449e-07, + "loss": 1.8599, + "step": 3 + }, + { + "epoch": 0.0009507962918944616, + "grad_norm": 116.69074877796281, + "learning_rate": 3.149606299212599e-07, + "loss": 1.0291, + "step": 4 + }, + { + "epoch": 0.001188495364868077, + "grad_norm": 1089.3112264958181, + "learning_rate": 3.937007874015748e-07, + "loss": 1.1106, + "step": 5 + }, + { + "epoch": 0.0014261944378416924, + "grad_norm": 278.54494736092164, + "learning_rate": 4.724409448818898e-07, + "loss": 1.031, + "step": 6 + }, + { + "epoch": 0.0016638935108153079, + "grad_norm": 880.929294966225, + "learning_rate": 5.511811023622048e-07, + "loss": 1.1007, + "step": 7 + }, + { + "epoch": 0.0019015925837889231, + "grad_norm": 3.9749308722917043, + "learning_rate": 6.299212598425198e-07, + "loss": 1.0455, + "step": 8 + }, + { + "epoch": 0.0021392916567625386, + "grad_norm": 9.616820257447039, + "learning_rate": 7.086614173228346e-07, + "loss": 1.0714, + "step": 9 + }, + { + "epoch": 0.002376990729736154, + "grad_norm": 9.110714766773242, + "learning_rate": 7.874015748031496e-07, + "loss": 1.0425, + "step": 10 + }, + { + "epoch": 0.002614689802709769, + "grad_norm": 35.47419288603955, + "learning_rate": 8.661417322834646e-07, + "loss": 1.0079, + "step": 11 + }, + { + "epoch": 0.0028523888756833847, + "grad_norm": 20.039055471812055, + "learning_rate": 9.448818897637796e-07, + "loss": 1.0568, + "step": 12 + }, + { + "epoch": 0.0030900879486570002, + "grad_norm": 97.91661458630223, + "learning_rate": 1.0236220472440946e-06, + "loss": 1.7756, + "step": 13 + }, + { + "epoch": 0.0033277870216306157, + "grad_norm": 280.99348737357474, + "learning_rate": 1.1023622047244096e-06, + "loss": 1.8025, + "step": 14 + }, + { + "epoch": 0.0035654860946042312, + "grad_norm": 14.415531454678254, + "learning_rate": 1.1811023622047246e-06, + "loss": 1.0604, + "step": 15 + }, + { + "epoch": 0.0038031851675778463, + "grad_norm": 3.8443210391372644, + "learning_rate": 1.2598425196850396e-06, + "loss": 0.9938, + "step": 16 + }, + { + "epoch": 0.004040884240551462, + "grad_norm": 4.672742446430964, + "learning_rate": 1.3385826771653545e-06, + "loss": 0.9687, + "step": 17 + }, + { + "epoch": 0.004278583313525077, + "grad_norm": 3.836449796292113, + "learning_rate": 1.4173228346456693e-06, + "loss": 0.9796, + "step": 18 + }, + { + "epoch": 0.004516282386498692, + "grad_norm": 158.0666830716214, + "learning_rate": 1.4960629921259845e-06, + "loss": 0.9845, + "step": 19 + }, + { + "epoch": 0.004753981459472308, + "grad_norm": 30.804480760358878, + "learning_rate": 1.5748031496062992e-06, + "loss": 0.962, + "step": 20 + }, + { + "epoch": 0.004991680532445923, + "grad_norm": 194.09125849047626, + "learning_rate": 1.6535433070866144e-06, + "loss": 1.7061, + "step": 21 + }, + { + "epoch": 0.005229379605419538, + "grad_norm": 2.9989843610593083, + "learning_rate": 1.7322834645669292e-06, + "loss": 0.9431, + "step": 22 + }, + { + "epoch": 0.005467078678393154, + "grad_norm": 3.240144565102094, + "learning_rate": 1.8110236220472444e-06, + "loss": 0.9823, + "step": 23 + }, + { + "epoch": 0.0057047777513667694, + "grad_norm": 32.58404786498518, + "learning_rate": 1.8897637795275591e-06, + "loss": 0.889, + "step": 24 + }, + { + "epoch": 0.005942476824340385, + "grad_norm": 2.7025804737514383, + "learning_rate": 1.968503937007874e-06, + "loss": 0.9508, + "step": 25 + }, + { + "epoch": 0.0061801758973140004, + "grad_norm": 3.381765986853288, + "learning_rate": 2.0472440944881893e-06, + "loss": 0.9111, + "step": 26 + }, + { + "epoch": 0.0064178749702876155, + "grad_norm": 3.6973303137941005, + "learning_rate": 2.125984251968504e-06, + "loss": 0.8947, + "step": 27 + }, + { + "epoch": 0.0066555740432612314, + "grad_norm": 2.670542810690506, + "learning_rate": 2.2047244094488192e-06, + "loss": 0.8229, + "step": 28 + }, + { + "epoch": 0.0068932731162348465, + "grad_norm": 3.9918119357193254, + "learning_rate": 2.283464566929134e-06, + "loss": 0.8771, + "step": 29 + }, + { + "epoch": 0.0071309721892084624, + "grad_norm": 4.5360983599057185, + "learning_rate": 2.362204724409449e-06, + "loss": 0.8642, + "step": 30 + }, + { + "epoch": 0.0073686712621820775, + "grad_norm": 2.112651282210059, + "learning_rate": 2.440944881889764e-06, + "loss": 0.7958, + "step": 31 + }, + { + "epoch": 0.007606370335155693, + "grad_norm": 42.98948966239728, + "learning_rate": 2.519685039370079e-06, + "loss": 0.8308, + "step": 32 + }, + { + "epoch": 0.007844069408129309, + "grad_norm": 2.8898285228603795, + "learning_rate": 2.598425196850394e-06, + "loss": 0.8072, + "step": 33 + }, + { + "epoch": 0.008081768481102924, + "grad_norm": 2.6990123318776527, + "learning_rate": 2.677165354330709e-06, + "loss": 0.7712, + "step": 34 + }, + { + "epoch": 0.008319467554076539, + "grad_norm": 3.39845149980693, + "learning_rate": 2.755905511811024e-06, + "loss": 0.8501, + "step": 35 + }, + { + "epoch": 0.008557166627050155, + "grad_norm": 2.2983980400400466, + "learning_rate": 2.8346456692913386e-06, + "loss": 0.817, + "step": 36 + }, + { + "epoch": 0.00879486570002377, + "grad_norm": 2.6568922854471597, + "learning_rate": 2.9133858267716538e-06, + "loss": 0.8167, + "step": 37 + }, + { + "epoch": 0.009032564772997385, + "grad_norm": 1.6900545589771088, + "learning_rate": 2.992125984251969e-06, + "loss": 0.8105, + "step": 38 + }, + { + "epoch": 0.009270263845971, + "grad_norm": 8.13453880713238, + "learning_rate": 3.0708661417322837e-06, + "loss": 0.8695, + "step": 39 + }, + { + "epoch": 0.009507962918944617, + "grad_norm": 2.0775870168573025, + "learning_rate": 3.1496062992125985e-06, + "loss": 0.8118, + "step": 40 + }, + { + "epoch": 0.00974566199191823, + "grad_norm": 1.5330267964160933, + "learning_rate": 3.2283464566929136e-06, + "loss": 0.8592, + "step": 41 + }, + { + "epoch": 0.009983361064891847, + "grad_norm": 6.2945191165404015, + "learning_rate": 3.307086614173229e-06, + "loss": 0.7847, + "step": 42 + }, + { + "epoch": 0.010221060137865463, + "grad_norm": 1.4100663998784049, + "learning_rate": 3.3858267716535436e-06, + "loss": 0.786, + "step": 43 + }, + { + "epoch": 0.010458759210839077, + "grad_norm": 1.681212061367187, + "learning_rate": 3.4645669291338583e-06, + "loss": 0.8168, + "step": 44 + }, + { + "epoch": 0.010696458283812693, + "grad_norm": 1.491170841282434, + "learning_rate": 3.5433070866141735e-06, + "loss": 0.8132, + "step": 45 + }, + { + "epoch": 0.010934157356786309, + "grad_norm": 1.5143707967966142, + "learning_rate": 3.6220472440944887e-06, + "loss": 0.8298, + "step": 46 + }, + { + "epoch": 0.011171856429759925, + "grad_norm": 1.3529518671003706, + "learning_rate": 3.7007874015748035e-06, + "loss": 0.7829, + "step": 47 + }, + { + "epoch": 0.011409555502733539, + "grad_norm": 37.75884344676398, + "learning_rate": 3.7795275590551182e-06, + "loss": 0.7701, + "step": 48 + }, + { + "epoch": 0.011647254575707155, + "grad_norm": 1.331111298859677, + "learning_rate": 3.858267716535433e-06, + "loss": 0.7913, + "step": 49 + }, + { + "epoch": 0.01188495364868077, + "grad_norm": 1.961075699947444, + "learning_rate": 3.937007874015748e-06, + "loss": 0.7282, + "step": 50 + }, + { + "epoch": 0.012122652721654385, + "grad_norm": 1.379073339064801, + "learning_rate": 4.015748031496064e-06, + "loss": 0.7762, + "step": 51 + }, + { + "epoch": 0.012360351794628001, + "grad_norm": 1.578516231836325, + "learning_rate": 4.0944881889763785e-06, + "loss": 0.8431, + "step": 52 + }, + { + "epoch": 0.012598050867601617, + "grad_norm": 1.683896282300194, + "learning_rate": 4.173228346456693e-06, + "loss": 0.7284, + "step": 53 + }, + { + "epoch": 0.012835749940575231, + "grad_norm": 1.3107959251346815, + "learning_rate": 4.251968503937008e-06, + "loss": 0.7609, + "step": 54 + }, + { + "epoch": 0.013073449013548847, + "grad_norm": 1.3544063514815377, + "learning_rate": 4.330708661417324e-06, + "loss": 0.7567, + "step": 55 + }, + { + "epoch": 0.013311148086522463, + "grad_norm": 1.6144825176592843, + "learning_rate": 4.4094488188976384e-06, + "loss": 0.7129, + "step": 56 + }, + { + "epoch": 0.013548847159496079, + "grad_norm": 1.289546605357833, + "learning_rate": 4.488188976377953e-06, + "loss": 0.7451, + "step": 57 + }, + { + "epoch": 0.013786546232469693, + "grad_norm": 1.2451927948654582, + "learning_rate": 4.566929133858268e-06, + "loss": 0.7704, + "step": 58 + }, + { + "epoch": 0.014024245305443309, + "grad_norm": 1.2635643758726904, + "learning_rate": 4.645669291338583e-06, + "loss": 0.731, + "step": 59 + }, + { + "epoch": 0.014261944378416925, + "grad_norm": 1.6281692523248195, + "learning_rate": 4.724409448818898e-06, + "loss": 0.7936, + "step": 60 + }, + { + "epoch": 0.014499643451390539, + "grad_norm": 1.2721216669227062, + "learning_rate": 4.803149606299213e-06, + "loss": 0.7754, + "step": 61 + }, + { + "epoch": 0.014737342524364155, + "grad_norm": 1.3563894389470352, + "learning_rate": 4.881889763779528e-06, + "loss": 0.7184, + "step": 62 + }, + { + "epoch": 0.014975041597337771, + "grad_norm": 1.2135038109928342, + "learning_rate": 4.960629921259843e-06, + "loss": 0.785, + "step": 63 + }, + { + "epoch": 0.015212740670311385, + "grad_norm": 1.458910152640145, + "learning_rate": 5.039370078740158e-06, + "loss": 0.7209, + "step": 64 + }, + { + "epoch": 0.015450439743285001, + "grad_norm": 1.3365676950147594, + "learning_rate": 5.118110236220473e-06, + "loss": 0.7594, + "step": 65 + }, + { + "epoch": 0.015688138816258617, + "grad_norm": 1.4643592734782072, + "learning_rate": 5.196850393700788e-06, + "loss": 0.7434, + "step": 66 + }, + { + "epoch": 0.01592583788923223, + "grad_norm": 1.4177782337993619, + "learning_rate": 5.2755905511811025e-06, + "loss": 0.7004, + "step": 67 + }, + { + "epoch": 0.01616353696220585, + "grad_norm": 1.412473167903132, + "learning_rate": 5.354330708661418e-06, + "loss": 0.7417, + "step": 68 + }, + { + "epoch": 0.016401236035179463, + "grad_norm": 1.4685482462527006, + "learning_rate": 5.433070866141733e-06, + "loss": 0.6792, + "step": 69 + }, + { + "epoch": 0.016638935108153077, + "grad_norm": 1.393003585454628, + "learning_rate": 5.511811023622048e-06, + "loss": 0.7232, + "step": 70 + }, + { + "epoch": 0.016876634181126695, + "grad_norm": 1.4386781851266082, + "learning_rate": 5.590551181102362e-06, + "loss": 0.748, + "step": 71 + }, + { + "epoch": 0.01711433325410031, + "grad_norm": 1.766128004078512, + "learning_rate": 5.669291338582677e-06, + "loss": 0.7488, + "step": 72 + }, + { + "epoch": 0.017352032327073923, + "grad_norm": 1.4847870328638093, + "learning_rate": 5.748031496062993e-06, + "loss": 0.7276, + "step": 73 + }, + { + "epoch": 0.01758973140004754, + "grad_norm": 1.3985746518747615, + "learning_rate": 5.8267716535433075e-06, + "loss": 0.7236, + "step": 74 + }, + { + "epoch": 0.017827430473021155, + "grad_norm": 1.4503002234063387, + "learning_rate": 5.905511811023622e-06, + "loss": 0.7386, + "step": 75 + }, + { + "epoch": 0.01806512954599477, + "grad_norm": 1.2418835083781392, + "learning_rate": 5.984251968503938e-06, + "loss": 0.7462, + "step": 76 + }, + { + "epoch": 0.018302828618968387, + "grad_norm": 1.6905950784160593, + "learning_rate": 6.062992125984253e-06, + "loss": 0.7803, + "step": 77 + }, + { + "epoch": 0.018540527691942, + "grad_norm": 1.3450987399339571, + "learning_rate": 6.141732283464567e-06, + "loss": 0.721, + "step": 78 + }, + { + "epoch": 0.018778226764915616, + "grad_norm": 1.1899703889469897, + "learning_rate": 6.220472440944882e-06, + "loss": 0.6856, + "step": 79 + }, + { + "epoch": 0.019015925837889233, + "grad_norm": 1.445717475630837, + "learning_rate": 6.299212598425197e-06, + "loss": 0.7075, + "step": 80 + }, + { + "epoch": 0.019253624910862847, + "grad_norm": 1.318781510485274, + "learning_rate": 6.3779527559055125e-06, + "loss": 0.7034, + "step": 81 + }, + { + "epoch": 0.01949132398383646, + "grad_norm": 1.2915827928180026, + "learning_rate": 6.456692913385827e-06, + "loss": 0.6647, + "step": 82 + }, + { + "epoch": 0.01972902305681008, + "grad_norm": 1.3965813486437952, + "learning_rate": 6.535433070866142e-06, + "loss": 0.7959, + "step": 83 + }, + { + "epoch": 0.019966722129783693, + "grad_norm": 1.5100549521348756, + "learning_rate": 6.614173228346458e-06, + "loss": 0.7153, + "step": 84 + }, + { + "epoch": 0.020204421202757308, + "grad_norm": 1.5028547161252093, + "learning_rate": 6.692913385826772e-06, + "loss": 0.7138, + "step": 85 + }, + { + "epoch": 0.020442120275730925, + "grad_norm": 2.0624361825362536, + "learning_rate": 6.771653543307087e-06, + "loss": 0.74, + "step": 86 + }, + { + "epoch": 0.02067981934870454, + "grad_norm": 1.1917986785150116, + "learning_rate": 6.850393700787402e-06, + "loss": 0.7002, + "step": 87 + }, + { + "epoch": 0.020917518421678154, + "grad_norm": 1.363229303251041, + "learning_rate": 6.929133858267717e-06, + "loss": 0.7415, + "step": 88 + }, + { + "epoch": 0.02115521749465177, + "grad_norm": 1.2824970167619907, + "learning_rate": 7.0078740157480315e-06, + "loss": 0.7262, + "step": 89 + }, + { + "epoch": 0.021392916567625386, + "grad_norm": 1.1767090182517193, + "learning_rate": 7.086614173228347e-06, + "loss": 0.6946, + "step": 90 + }, + { + "epoch": 0.021630615640599003, + "grad_norm": 1.7243211552918418, + "learning_rate": 7.165354330708662e-06, + "loss": 0.7641, + "step": 91 + }, + { + "epoch": 0.021868314713572617, + "grad_norm": 1.3669395448614943, + "learning_rate": 7.2440944881889774e-06, + "loss": 0.6917, + "step": 92 + }, + { + "epoch": 0.02210601378654623, + "grad_norm": 1.4947395516690123, + "learning_rate": 7.322834645669292e-06, + "loss": 0.7555, + "step": 93 + }, + { + "epoch": 0.02234371285951985, + "grad_norm": 1.3130032808162058, + "learning_rate": 7.401574803149607e-06, + "loss": 0.6952, + "step": 94 + }, + { + "epoch": 0.022581411932493464, + "grad_norm": 1.364158007005569, + "learning_rate": 7.480314960629922e-06, + "loss": 0.6111, + "step": 95 + }, + { + "epoch": 0.022819111005467078, + "grad_norm": 1.287280230131665, + "learning_rate": 7.5590551181102365e-06, + "loss": 0.6963, + "step": 96 + }, + { + "epoch": 0.023056810078440695, + "grad_norm": 1.3300314350924267, + "learning_rate": 7.637795275590551e-06, + "loss": 0.7527, + "step": 97 + }, + { + "epoch": 0.02329450915141431, + "grad_norm": 1.2655028735521923, + "learning_rate": 7.716535433070867e-06, + "loss": 0.7353, + "step": 98 + }, + { + "epoch": 0.023532208224387924, + "grad_norm": 1.615528897060331, + "learning_rate": 7.79527559055118e-06, + "loss": 0.7464, + "step": 99 + }, + { + "epoch": 0.02376990729736154, + "grad_norm": 1.365013875313374, + "learning_rate": 7.874015748031496e-06, + "loss": 0.7234, + "step": 100 + }, + { + "epoch": 0.02376990729736154, + "eval_loss": 0.7271476984024048, + "eval_runtime": 295.9386, + "eval_samples_per_second": 36.758, + "eval_steps_per_second": 1.149, + "step": 100 + }, + { + "epoch": 0.024007606370335156, + "grad_norm": 1.3718807914215128, + "learning_rate": 7.952755905511812e-06, + "loss": 0.7412, + "step": 101 + }, + { + "epoch": 0.02424530544330877, + "grad_norm": 1.2268223225586974, + "learning_rate": 8.031496062992128e-06, + "loss": 0.7717, + "step": 102 + }, + { + "epoch": 0.024483004516282388, + "grad_norm": 1.2906683544994684, + "learning_rate": 8.110236220472441e-06, + "loss": 0.7205, + "step": 103 + }, + { + "epoch": 0.024720703589256002, + "grad_norm": 2.2950196223000097, + "learning_rate": 8.188976377952757e-06, + "loss": 0.7304, + "step": 104 + }, + { + "epoch": 0.024958402662229616, + "grad_norm": 1.4437313124339044, + "learning_rate": 8.267716535433071e-06, + "loss": 0.7158, + "step": 105 + }, + { + "epoch": 0.025196101735203234, + "grad_norm": 1.4224478878431277, + "learning_rate": 8.346456692913387e-06, + "loss": 0.7347, + "step": 106 + }, + { + "epoch": 0.025433800808176848, + "grad_norm": 1.43746814853937, + "learning_rate": 8.4251968503937e-06, + "loss": 0.7138, + "step": 107 + }, + { + "epoch": 0.025671499881150462, + "grad_norm": 1.3555381341309354, + "learning_rate": 8.503937007874016e-06, + "loss": 0.7269, + "step": 108 + }, + { + "epoch": 0.02590919895412408, + "grad_norm": 1.6659281695284676, + "learning_rate": 8.582677165354332e-06, + "loss": 0.7845, + "step": 109 + }, + { + "epoch": 0.026146898027097694, + "grad_norm": 1.202922625412947, + "learning_rate": 8.661417322834647e-06, + "loss": 0.645, + "step": 110 + }, + { + "epoch": 0.026384597100071308, + "grad_norm": 1.3838131254595956, + "learning_rate": 8.740157480314961e-06, + "loss": 0.7027, + "step": 111 + }, + { + "epoch": 0.026622296173044926, + "grad_norm": 1.5330680808908475, + "learning_rate": 8.818897637795277e-06, + "loss": 0.6497, + "step": 112 + }, + { + "epoch": 0.02685999524601854, + "grad_norm": 1.524635438800432, + "learning_rate": 8.89763779527559e-06, + "loss": 0.7005, + "step": 113 + }, + { + "epoch": 0.027097694318992158, + "grad_norm": 2.060883647727988, + "learning_rate": 8.976377952755906e-06, + "loss": 0.6847, + "step": 114 + }, + { + "epoch": 0.027335393391965772, + "grad_norm": 2.833246418056645, + "learning_rate": 9.05511811023622e-06, + "loss": 0.706, + "step": 115 + }, + { + "epoch": 0.027573092464939386, + "grad_norm": 1.6420643802510664, + "learning_rate": 9.133858267716536e-06, + "loss": 0.777, + "step": 116 + }, + { + "epoch": 0.027810791537913004, + "grad_norm": 1.9107284666643607, + "learning_rate": 9.212598425196852e-06, + "loss": 0.6882, + "step": 117 + }, + { + "epoch": 0.028048490610886618, + "grad_norm": 1.2727092079528393, + "learning_rate": 9.291338582677165e-06, + "loss": 0.6807, + "step": 118 + }, + { + "epoch": 0.028286189683860232, + "grad_norm": 1.1940424934666167, + "learning_rate": 9.370078740157481e-06, + "loss": 0.7567, + "step": 119 + }, + { + "epoch": 0.02852388875683385, + "grad_norm": 1.345325215182601, + "learning_rate": 9.448818897637797e-06, + "loss": 0.7182, + "step": 120 + }, + { + "epoch": 0.028761587829807464, + "grad_norm": 1.2739117565353772, + "learning_rate": 9.52755905511811e-06, + "loss": 0.7267, + "step": 121 + }, + { + "epoch": 0.028999286902781078, + "grad_norm": 1.809527576574249, + "learning_rate": 9.606299212598426e-06, + "loss": 0.7278, + "step": 122 + }, + { + "epoch": 0.029236985975754696, + "grad_norm": 2.428680696726001, + "learning_rate": 9.68503937007874e-06, + "loss": 0.6907, + "step": 123 + }, + { + "epoch": 0.02947468504872831, + "grad_norm": 1.3936655044512958, + "learning_rate": 9.763779527559056e-06, + "loss": 0.6811, + "step": 124 + }, + { + "epoch": 0.029712384121701924, + "grad_norm": 4.6236743595202405, + "learning_rate": 9.842519685039371e-06, + "loss": 0.7529, + "step": 125 + }, + { + "epoch": 0.029950083194675542, + "grad_norm": 1.2613707416510274, + "learning_rate": 9.921259842519685e-06, + "loss": 0.6759, + "step": 126 + }, + { + "epoch": 0.030187782267649156, + "grad_norm": 1.1612183632433257, + "learning_rate": 1e-05, + "loss": 0.7322, + "step": 127 + }, + { + "epoch": 0.03042548134062277, + "grad_norm": 1.1975284871985445, + "learning_rate": 9.999998517757006e-06, + "loss": 0.7415, + "step": 128 + }, + { + "epoch": 0.030663180413596388, + "grad_norm": 1.2579909616847225, + "learning_rate": 9.999994071028903e-06, + "loss": 0.6749, + "step": 129 + }, + { + "epoch": 0.030900879486570002, + "grad_norm": 1.7774117338298603, + "learning_rate": 9.999986659818324e-06, + "loss": 0.7511, + "step": 130 + }, + { + "epoch": 0.031138578559543616, + "grad_norm": 2.2229354280424074, + "learning_rate": 9.999976284129664e-06, + "loss": 0.7167, + "step": 131 + }, + { + "epoch": 0.031376277632517234, + "grad_norm": 1.766695808522094, + "learning_rate": 9.99996294396908e-06, + "loss": 0.6882, + "step": 132 + }, + { + "epoch": 0.03161397670549085, + "grad_norm": 1.3989756991019069, + "learning_rate": 9.999946639344475e-06, + "loss": 0.6878, + "step": 133 + }, + { + "epoch": 0.03185167577846446, + "grad_norm": 1.3384859934369597, + "learning_rate": 9.999927370265519e-06, + "loss": 0.6804, + "step": 134 + }, + { + "epoch": 0.03208937485143808, + "grad_norm": 1.3994593115221146, + "learning_rate": 9.999905136743635e-06, + "loss": 0.7172, + "step": 135 + }, + { + "epoch": 0.0323270739244117, + "grad_norm": 1.2639889957252497, + "learning_rate": 9.999879938792008e-06, + "loss": 0.7298, + "step": 136 + }, + { + "epoch": 0.03256477299738531, + "grad_norm": 1.2307123508701523, + "learning_rate": 9.999851776425575e-06, + "loss": 0.7122, + "step": 137 + }, + { + "epoch": 0.032802472070358926, + "grad_norm": 1.2290582011237643, + "learning_rate": 9.999820649661033e-06, + "loss": 0.7328, + "step": 138 + }, + { + "epoch": 0.03304017114333254, + "grad_norm": 1.572379217478301, + "learning_rate": 9.99978655851684e-06, + "loss": 0.7451, + "step": 139 + }, + { + "epoch": 0.033277870216306155, + "grad_norm": 1.2847030013677379, + "learning_rate": 9.999749503013207e-06, + "loss": 0.7066, + "step": 140 + }, + { + "epoch": 0.03351556928927977, + "grad_norm": 1.2593996962352485, + "learning_rate": 9.999709483172103e-06, + "loss": 0.7131, + "step": 141 + }, + { + "epoch": 0.03375326836225339, + "grad_norm": 1.684581623326016, + "learning_rate": 9.999666499017257e-06, + "loss": 0.683, + "step": 142 + }, + { + "epoch": 0.033990967435227004, + "grad_norm": 1.386908432598645, + "learning_rate": 9.999620550574155e-06, + "loss": 0.7588, + "step": 143 + }, + { + "epoch": 0.03422866650820062, + "grad_norm": 1.2700253347892168, + "learning_rate": 9.999571637870035e-06, + "loss": 0.6374, + "step": 144 + }, + { + "epoch": 0.03446636558117423, + "grad_norm": 1.2026629545066871, + "learning_rate": 9.999519760933905e-06, + "loss": 0.6913, + "step": 145 + }, + { + "epoch": 0.03470406465414785, + "grad_norm": 1.9534124469127874, + "learning_rate": 9.999464919796516e-06, + "loss": 0.7686, + "step": 146 + }, + { + "epoch": 0.03494176372712146, + "grad_norm": 1.8018782959112754, + "learning_rate": 9.999407114490384e-06, + "loss": 0.6715, + "step": 147 + }, + { + "epoch": 0.03517946280009508, + "grad_norm": 1.2670615091688917, + "learning_rate": 9.999346345049786e-06, + "loss": 0.6953, + "step": 148 + }, + { + "epoch": 0.035417161873068696, + "grad_norm": 1.4400217651478113, + "learning_rate": 9.999282611510749e-06, + "loss": 0.7996, + "step": 149 + }, + { + "epoch": 0.03565486094604231, + "grad_norm": 1.510016539355765, + "learning_rate": 9.99921591391106e-06, + "loss": 0.7466, + "step": 150 + }, + { + "epoch": 0.035892560019015925, + "grad_norm": 1.2796172646461035, + "learning_rate": 9.999146252290264e-06, + "loss": 0.6511, + "step": 151 + }, + { + "epoch": 0.03613025909198954, + "grad_norm": 1.2095545464132496, + "learning_rate": 9.999073626689664e-06, + "loss": 0.7183, + "step": 152 + }, + { + "epoch": 0.03636795816496316, + "grad_norm": 1.4068698809502262, + "learning_rate": 9.998998037152318e-06, + "loss": 0.7321, + "step": 153 + }, + { + "epoch": 0.036605657237936774, + "grad_norm": 1.3466005961281857, + "learning_rate": 9.998919483723046e-06, + "loss": 0.6645, + "step": 154 + }, + { + "epoch": 0.03684335631091039, + "grad_norm": 1.8917390770664444, + "learning_rate": 9.99883796644842e-06, + "loss": 0.7173, + "step": 155 + }, + { + "epoch": 0.037081055383884, + "grad_norm": 1.3371402910011143, + "learning_rate": 9.998753485376772e-06, + "loss": 0.6802, + "step": 156 + }, + { + "epoch": 0.03731875445685762, + "grad_norm": 1.8482124183300186, + "learning_rate": 9.998666040558187e-06, + "loss": 0.7325, + "step": 157 + }, + { + "epoch": 0.03755645352983123, + "grad_norm": 1.711904492084812, + "learning_rate": 9.998575632044516e-06, + "loss": 0.7301, + "step": 158 + }, + { + "epoch": 0.03779415260280485, + "grad_norm": 1.4646386224697934, + "learning_rate": 9.99848225988936e-06, + "loss": 0.728, + "step": 159 + }, + { + "epoch": 0.038031851675778466, + "grad_norm": 1.300872582898599, + "learning_rate": 9.998385924148077e-06, + "loss": 0.7039, + "step": 160 + }, + { + "epoch": 0.03826955074875208, + "grad_norm": 1.2690798893892725, + "learning_rate": 9.998286624877786e-06, + "loss": 0.7039, + "step": 161 + }, + { + "epoch": 0.038507249821725695, + "grad_norm": 1.3530938939510513, + "learning_rate": 9.998184362137364e-06, + "loss": 0.7109, + "step": 162 + }, + { + "epoch": 0.03874494889469931, + "grad_norm": 1.4574778143625637, + "learning_rate": 9.998079135987437e-06, + "loss": 0.6798, + "step": 163 + }, + { + "epoch": 0.03898264796767292, + "grad_norm": 1.294422043613207, + "learning_rate": 9.997970946490398e-06, + "loss": 0.6945, + "step": 164 + }, + { + "epoch": 0.039220347040646544, + "grad_norm": 1.4658185974242126, + "learning_rate": 9.99785979371039e-06, + "loss": 0.7582, + "step": 165 + }, + { + "epoch": 0.03945804611362016, + "grad_norm": 1.277305750459812, + "learning_rate": 9.997745677713313e-06, + "loss": 0.7085, + "step": 166 + }, + { + "epoch": 0.03969574518659377, + "grad_norm": 1.532523260851834, + "learning_rate": 9.99762859856683e-06, + "loss": 0.6875, + "step": 167 + }, + { + "epoch": 0.03993344425956739, + "grad_norm": 1.5631555156805317, + "learning_rate": 9.997508556340354e-06, + "loss": 0.7598, + "step": 168 + }, + { + "epoch": 0.040171143332541, + "grad_norm": 1.1745658999606259, + "learning_rate": 9.997385551105061e-06, + "loss": 0.6547, + "step": 169 + }, + { + "epoch": 0.040408842405514615, + "grad_norm": 1.19597097984527, + "learning_rate": 9.997259582933878e-06, + "loss": 0.639, + "step": 170 + }, + { + "epoch": 0.040646541478488236, + "grad_norm": 1.4486469664588677, + "learning_rate": 9.997130651901491e-06, + "loss": 0.687, + "step": 171 + }, + { + "epoch": 0.04088424055146185, + "grad_norm": 1.3548957989463, + "learning_rate": 9.996998758084344e-06, + "loss": 0.7696, + "step": 172 + }, + { + "epoch": 0.041121939624435465, + "grad_norm": 1.3056235637781557, + "learning_rate": 9.996863901560635e-06, + "loss": 0.7053, + "step": 173 + }, + { + "epoch": 0.04135963869740908, + "grad_norm": 1.2969600488947626, + "learning_rate": 9.996726082410322e-06, + "loss": 0.6942, + "step": 174 + }, + { + "epoch": 0.04159733777038269, + "grad_norm": 1.236388269561959, + "learning_rate": 9.996585300715117e-06, + "loss": 0.6827, + "step": 175 + }, + { + "epoch": 0.04183503684335631, + "grad_norm": 1.6481540118145737, + "learning_rate": 9.996441556558485e-06, + "loss": 0.7051, + "step": 176 + }, + { + "epoch": 0.04207273591632993, + "grad_norm": 1.5068724698139169, + "learning_rate": 9.996294850025658e-06, + "loss": 0.7212, + "step": 177 + }, + { + "epoch": 0.04231043498930354, + "grad_norm": 1.8865145758678292, + "learning_rate": 9.996145181203616e-06, + "loss": 0.7236, + "step": 178 + }, + { + "epoch": 0.04254813406227716, + "grad_norm": 2.935356619908941, + "learning_rate": 9.995992550181095e-06, + "loss": 0.6606, + "step": 179 + }, + { + "epoch": 0.04278583313525077, + "grad_norm": 1.5260870100050417, + "learning_rate": 9.99583695704859e-06, + "loss": 0.7063, + "step": 180 + }, + { + "epoch": 0.043023532208224385, + "grad_norm": 1.4751286429941906, + "learning_rate": 9.995678401898354e-06, + "loss": 0.7288, + "step": 181 + }, + { + "epoch": 0.04326123128119801, + "grad_norm": 1.1492281892125629, + "learning_rate": 9.995516884824393e-06, + "loss": 0.6619, + "step": 182 + }, + { + "epoch": 0.04349893035417162, + "grad_norm": 1.8150829427785264, + "learning_rate": 9.995352405922467e-06, + "loss": 0.7226, + "step": 183 + }, + { + "epoch": 0.043736629427145235, + "grad_norm": 1.516070573517932, + "learning_rate": 9.995184965290099e-06, + "loss": 0.7494, + "step": 184 + }, + { + "epoch": 0.04397432850011885, + "grad_norm": 1.2622881989225228, + "learning_rate": 9.995014563026561e-06, + "loss": 0.7392, + "step": 185 + }, + { + "epoch": 0.04421202757309246, + "grad_norm": 1.376599894473054, + "learning_rate": 9.994841199232888e-06, + "loss": 0.6844, + "step": 186 + }, + { + "epoch": 0.04444972664606608, + "grad_norm": 1.374367997022601, + "learning_rate": 9.994664874011864e-06, + "loss": 0.738, + "step": 187 + }, + { + "epoch": 0.0446874257190397, + "grad_norm": 1.4910974555653231, + "learning_rate": 9.99448558746803e-06, + "loss": 0.7708, + "step": 188 + }, + { + "epoch": 0.04492512479201331, + "grad_norm": 1.415053873756753, + "learning_rate": 9.99430333970769e-06, + "loss": 0.7194, + "step": 189 + }, + { + "epoch": 0.04516282386498693, + "grad_norm": 1.3227011816044711, + "learning_rate": 9.994118130838893e-06, + "loss": 0.6975, + "step": 190 + }, + { + "epoch": 0.04540052293796054, + "grad_norm": 1.9907210442742524, + "learning_rate": 9.99392996097145e-06, + "loss": 0.6784, + "step": 191 + }, + { + "epoch": 0.045638222010934155, + "grad_norm": 1.7458519012917117, + "learning_rate": 9.993738830216929e-06, + "loss": 0.7039, + "step": 192 + }, + { + "epoch": 0.04587592108390777, + "grad_norm": 1.5348288990153172, + "learning_rate": 9.993544738688647e-06, + "loss": 0.7131, + "step": 193 + }, + { + "epoch": 0.04611362015688139, + "grad_norm": 1.3208462005895787, + "learning_rate": 9.993347686501684e-06, + "loss": 0.7236, + "step": 194 + }, + { + "epoch": 0.046351319229855005, + "grad_norm": 1.4016725155886078, + "learning_rate": 9.993147673772869e-06, + "loss": 0.7159, + "step": 195 + }, + { + "epoch": 0.04658901830282862, + "grad_norm": 1.4163757619712334, + "learning_rate": 9.992944700620793e-06, + "loss": 0.6818, + "step": 196 + }, + { + "epoch": 0.04682671737580223, + "grad_norm": 1.4024158751349651, + "learning_rate": 9.992738767165791e-06, + "loss": 0.7045, + "step": 197 + }, + { + "epoch": 0.04706441644877585, + "grad_norm": 6.829868690412084, + "learning_rate": 9.992529873529966e-06, + "loss": 0.6322, + "step": 198 + }, + { + "epoch": 0.04730211552174946, + "grad_norm": 1.3824918447757246, + "learning_rate": 9.992318019837171e-06, + "loss": 0.7352, + "step": 199 + }, + { + "epoch": 0.04753981459472308, + "grad_norm": 1.1407460584291533, + "learning_rate": 9.992103206213009e-06, + "loss": 0.6613, + "step": 200 + }, + { + "epoch": 0.04753981459472308, + "eval_loss": 0.7093249559402466, + "eval_runtime": 277.7068, + "eval_samples_per_second": 39.171, + "eval_steps_per_second": 1.224, + "step": 200 + }, + { + "epoch": 0.0477775136676967, + "grad_norm": 1.206156480098467, + "learning_rate": 9.991885432784848e-06, + "loss": 0.6413, + "step": 201 + }, + { + "epoch": 0.04801521274067031, + "grad_norm": 1.374108137610864, + "learning_rate": 9.9916646996818e-06, + "loss": 0.6882, + "step": 202 + }, + { + "epoch": 0.048252911813643926, + "grad_norm": 1.2536938122500216, + "learning_rate": 9.991441007034738e-06, + "loss": 0.686, + "step": 203 + }, + { + "epoch": 0.04849061088661754, + "grad_norm": 1.2803980178530157, + "learning_rate": 9.99121435497629e-06, + "loss": 0.6667, + "step": 204 + }, + { + "epoch": 0.04872830995959116, + "grad_norm": 1.510390765173908, + "learning_rate": 9.990984743640839e-06, + "loss": 0.7565, + "step": 205 + }, + { + "epoch": 0.048966009032564775, + "grad_norm": 2.6167368887196627, + "learning_rate": 9.99075217316452e-06, + "loss": 0.7031, + "step": 206 + }, + { + "epoch": 0.04920370810553839, + "grad_norm": 1.3950688327598426, + "learning_rate": 9.990516643685222e-06, + "loss": 0.6767, + "step": 207 + }, + { + "epoch": 0.049441407178512004, + "grad_norm": 1.377712342852086, + "learning_rate": 9.99027815534259e-06, + "loss": 0.7274, + "step": 208 + }, + { + "epoch": 0.04967910625148562, + "grad_norm": 1.2828995891134902, + "learning_rate": 9.990036708278023e-06, + "loss": 0.6673, + "step": 209 + }, + { + "epoch": 0.04991680532445923, + "grad_norm": 1.4098615455430532, + "learning_rate": 9.989792302634674e-06, + "loss": 0.6731, + "step": 210 + }, + { + "epoch": 0.05015450439743285, + "grad_norm": 1.3204289261081632, + "learning_rate": 9.989544938557453e-06, + "loss": 0.7185, + "step": 211 + }, + { + "epoch": 0.05039220347040647, + "grad_norm": 1.267272100548898, + "learning_rate": 9.989294616193018e-06, + "loss": 0.7075, + "step": 212 + }, + { + "epoch": 0.05062990254338008, + "grad_norm": 1.4240685143442575, + "learning_rate": 9.989041335689788e-06, + "loss": 0.6749, + "step": 213 + }, + { + "epoch": 0.050867601616353696, + "grad_norm": 1.3674794425537904, + "learning_rate": 9.988785097197927e-06, + "loss": 0.7179, + "step": 214 + }, + { + "epoch": 0.05110530068932731, + "grad_norm": 1.435810023419144, + "learning_rate": 9.988525900869366e-06, + "loss": 0.7205, + "step": 215 + }, + { + "epoch": 0.051342999762300924, + "grad_norm": 1.370554004421123, + "learning_rate": 9.988263746857773e-06, + "loss": 0.6602, + "step": 216 + }, + { + "epoch": 0.051580698835274545, + "grad_norm": 1.6303569992975493, + "learning_rate": 9.987998635318586e-06, + "loss": 0.6897, + "step": 217 + }, + { + "epoch": 0.05181839790824816, + "grad_norm": 1.2889397918036805, + "learning_rate": 9.987730566408983e-06, + "loss": 0.6874, + "step": 218 + }, + { + "epoch": 0.052056096981221774, + "grad_norm": 1.6183284461028713, + "learning_rate": 9.987459540287906e-06, + "loss": 0.6967, + "step": 219 + }, + { + "epoch": 0.05229379605419539, + "grad_norm": 1.3077089508296627, + "learning_rate": 9.987185557116042e-06, + "loss": 0.6901, + "step": 220 + }, + { + "epoch": 0.052531495127169, + "grad_norm": 1.2798437199584218, + "learning_rate": 9.986908617055836e-06, + "loss": 0.7543, + "step": 221 + }, + { + "epoch": 0.052769194200142616, + "grad_norm": 1.6151415724420985, + "learning_rate": 9.986628720271487e-06, + "loss": 0.7714, + "step": 222 + }, + { + "epoch": 0.05300689327311624, + "grad_norm": 1.0825175370035083, + "learning_rate": 9.98634586692894e-06, + "loss": 0.6627, + "step": 223 + }, + { + "epoch": 0.05324459234608985, + "grad_norm": 1.1991671286969896, + "learning_rate": 9.986060057195905e-06, + "loss": 0.6853, + "step": 224 + }, + { + "epoch": 0.053482291419063466, + "grad_norm": 1.215859644982796, + "learning_rate": 9.98577129124183e-06, + "loss": 0.7103, + "step": 225 + }, + { + "epoch": 0.05371999049203708, + "grad_norm": 1.646508314884775, + "learning_rate": 9.985479569237932e-06, + "loss": 0.7058, + "step": 226 + }, + { + "epoch": 0.053957689565010694, + "grad_norm": 1.3233648855680409, + "learning_rate": 9.985184891357165e-06, + "loss": 0.713, + "step": 227 + }, + { + "epoch": 0.054195388637984315, + "grad_norm": 1.5592220205015455, + "learning_rate": 9.984887257774246e-06, + "loss": 0.7234, + "step": 228 + }, + { + "epoch": 0.05443308771095793, + "grad_norm": 1.5568283854465528, + "learning_rate": 9.984586668665641e-06, + "loss": 0.7411, + "step": 229 + }, + { + "epoch": 0.054670786783931544, + "grad_norm": 1.7467025992167564, + "learning_rate": 9.984283124209567e-06, + "loss": 0.6796, + "step": 230 + }, + { + "epoch": 0.05490848585690516, + "grad_norm": 1.3266845733209982, + "learning_rate": 9.983976624585996e-06, + "loss": 0.5982, + "step": 231 + }, + { + "epoch": 0.05514618492987877, + "grad_norm": 1.6347103444762536, + "learning_rate": 9.983667169976651e-06, + "loss": 0.6859, + "step": 232 + }, + { + "epoch": 0.055383884002852386, + "grad_norm": 1.4856334735840893, + "learning_rate": 9.983354760565004e-06, + "loss": 0.7269, + "step": 233 + }, + { + "epoch": 0.05562158307582601, + "grad_norm": 1.2567450649039735, + "learning_rate": 9.983039396536286e-06, + "loss": 0.6933, + "step": 234 + }, + { + "epoch": 0.05585928214879962, + "grad_norm": 1.3675253014366846, + "learning_rate": 9.982721078077474e-06, + "loss": 0.6913, + "step": 235 + }, + { + "epoch": 0.056096981221773236, + "grad_norm": 1.216115234732168, + "learning_rate": 9.982399805377297e-06, + "loss": 0.6628, + "step": 236 + }, + { + "epoch": 0.05633468029474685, + "grad_norm": 1.2578638652616427, + "learning_rate": 9.982075578626235e-06, + "loss": 0.6381, + "step": 237 + }, + { + "epoch": 0.056572379367720464, + "grad_norm": 1.5020124396699994, + "learning_rate": 9.981748398016524e-06, + "loss": 0.6957, + "step": 238 + }, + { + "epoch": 0.05681007844069408, + "grad_norm": 1.3182476536037422, + "learning_rate": 9.981418263742148e-06, + "loss": 0.7457, + "step": 239 + }, + { + "epoch": 0.0570477775136677, + "grad_norm": 1.0763054374735448, + "learning_rate": 9.981085175998843e-06, + "loss": 0.64, + "step": 240 + }, + { + "epoch": 0.057285476586641314, + "grad_norm": 1.1661441013092808, + "learning_rate": 9.980749134984094e-06, + "loss": 0.6704, + "step": 241 + }, + { + "epoch": 0.05752317565961493, + "grad_norm": 1.5271965111821624, + "learning_rate": 9.98041014089714e-06, + "loss": 0.6829, + "step": 242 + }, + { + "epoch": 0.05776087473258854, + "grad_norm": 1.4464128574725144, + "learning_rate": 9.980068193938971e-06, + "loss": 0.7268, + "step": 243 + }, + { + "epoch": 0.057998573805562156, + "grad_norm": 1.265320604732494, + "learning_rate": 9.979723294312324e-06, + "loss": 0.773, + "step": 244 + }, + { + "epoch": 0.05823627287853577, + "grad_norm": 1.2736690889554387, + "learning_rate": 9.97937544222169e-06, + "loss": 0.7169, + "step": 245 + }, + { + "epoch": 0.05847397195150939, + "grad_norm": 1.2029863836615373, + "learning_rate": 9.979024637873309e-06, + "loss": 0.6604, + "step": 246 + }, + { + "epoch": 0.058711671024483006, + "grad_norm": 1.193644264340822, + "learning_rate": 9.978670881475173e-06, + "loss": 0.7295, + "step": 247 + }, + { + "epoch": 0.05894937009745662, + "grad_norm": 1.3011443121403328, + "learning_rate": 9.978314173237022e-06, + "loss": 0.7148, + "step": 248 + }, + { + "epoch": 0.059187069170430234, + "grad_norm": 1.248981307270402, + "learning_rate": 9.97795451337035e-06, + "loss": 0.6837, + "step": 249 + }, + { + "epoch": 0.05942476824340385, + "grad_norm": 1.268127870828261, + "learning_rate": 9.977591902088394e-06, + "loss": 0.7589, + "step": 250 + }, + { + "epoch": 0.05966246731637746, + "grad_norm": 1.3514532192217257, + "learning_rate": 9.977226339606149e-06, + "loss": 0.7547, + "step": 251 + }, + { + "epoch": 0.059900166389351084, + "grad_norm": 1.2791762606819672, + "learning_rate": 9.976857826140354e-06, + "loss": 0.7244, + "step": 252 + }, + { + "epoch": 0.0601378654623247, + "grad_norm": 1.2511065842896285, + "learning_rate": 9.9764863619095e-06, + "loss": 0.7413, + "step": 253 + }, + { + "epoch": 0.06037556453529831, + "grad_norm": 1.3016971558662096, + "learning_rate": 9.976111947133826e-06, + "loss": 0.7279, + "step": 254 + }, + { + "epoch": 0.060613263608271926, + "grad_norm": 1.8446744023734294, + "learning_rate": 9.975734582035323e-06, + "loss": 0.7112, + "step": 255 + }, + { + "epoch": 0.06085096268124554, + "grad_norm": 1.3772732715257179, + "learning_rate": 9.975354266837728e-06, + "loss": 0.6858, + "step": 256 + }, + { + "epoch": 0.06108866175421916, + "grad_norm": 1.1439033583883755, + "learning_rate": 9.974971001766534e-06, + "loss": 0.6875, + "step": 257 + }, + { + "epoch": 0.061326360827192776, + "grad_norm": 1.1751483264866174, + "learning_rate": 9.974584787048971e-06, + "loss": 0.6872, + "step": 258 + }, + { + "epoch": 0.06156405990016639, + "grad_norm": 1.2762690965668322, + "learning_rate": 9.97419562291403e-06, + "loss": 0.6992, + "step": 259 + }, + { + "epoch": 0.061801758973140004, + "grad_norm": 1.4093824623196907, + "learning_rate": 9.973803509592439e-06, + "loss": 0.6805, + "step": 260 + }, + { + "epoch": 0.06203945804611362, + "grad_norm": 1.1380406410292525, + "learning_rate": 9.973408447316689e-06, + "loss": 0.6952, + "step": 261 + }, + { + "epoch": 0.06227715711908723, + "grad_norm": 1.3117980662213509, + "learning_rate": 9.973010436321005e-06, + "loss": 0.6986, + "step": 262 + }, + { + "epoch": 0.06251485619206085, + "grad_norm": 1.1889048495633547, + "learning_rate": 9.972609476841368e-06, + "loss": 0.6537, + "step": 263 + }, + { + "epoch": 0.06275255526503447, + "grad_norm": 1.1614573269705215, + "learning_rate": 9.972205569115508e-06, + "loss": 0.688, + "step": 264 + }, + { + "epoch": 0.06299025433800808, + "grad_norm": 1.2491851607824525, + "learning_rate": 9.971798713382896e-06, + "loss": 0.7433, + "step": 265 + }, + { + "epoch": 0.0632279534109817, + "grad_norm": 1.4074800150180646, + "learning_rate": 9.971388909884763e-06, + "loss": 0.7296, + "step": 266 + }, + { + "epoch": 0.06346565248395532, + "grad_norm": 1.3189697687419493, + "learning_rate": 9.970976158864074e-06, + "loss": 0.7332, + "step": 267 + }, + { + "epoch": 0.06370335155692892, + "grad_norm": 1.2307915834627277, + "learning_rate": 9.97056046056555e-06, + "loss": 0.6693, + "step": 268 + }, + { + "epoch": 0.06394105062990255, + "grad_norm": 1.4258408660623567, + "learning_rate": 9.970141815235657e-06, + "loss": 0.7208, + "step": 269 + }, + { + "epoch": 0.06417874970287615, + "grad_norm": 1.91631091951768, + "learning_rate": 9.96972022312261e-06, + "loss": 0.6691, + "step": 270 + }, + { + "epoch": 0.06441644877584977, + "grad_norm": 1.3932567793327448, + "learning_rate": 9.96929568447637e-06, + "loss": 0.7239, + "step": 271 + }, + { + "epoch": 0.0646541478488234, + "grad_norm": 1.3406874043514843, + "learning_rate": 9.968868199548641e-06, + "loss": 0.6704, + "step": 272 + }, + { + "epoch": 0.064891846921797, + "grad_norm": 1.3185082004523028, + "learning_rate": 9.968437768592882e-06, + "loss": 0.7259, + "step": 273 + }, + { + "epoch": 0.06512954599477062, + "grad_norm": 1.3161013145338907, + "learning_rate": 9.968004391864292e-06, + "loss": 0.6757, + "step": 274 + }, + { + "epoch": 0.06536724506774423, + "grad_norm": 1.8478060859350474, + "learning_rate": 9.96756806961982e-06, + "loss": 0.7279, + "step": 275 + }, + { + "epoch": 0.06560494414071785, + "grad_norm": 1.3870123258910365, + "learning_rate": 9.96712880211816e-06, + "loss": 0.7206, + "step": 276 + }, + { + "epoch": 0.06584264321369146, + "grad_norm": 1.5656743825087394, + "learning_rate": 9.96668658961975e-06, + "loss": 0.6992, + "step": 277 + }, + { + "epoch": 0.06608034228666508, + "grad_norm": 1.299937726031343, + "learning_rate": 9.96624143238678e-06, + "loss": 0.7168, + "step": 278 + }, + { + "epoch": 0.0663180413596387, + "grad_norm": 1.3956906437606802, + "learning_rate": 9.965793330683182e-06, + "loss": 0.7284, + "step": 279 + }, + { + "epoch": 0.06655574043261231, + "grad_norm": 1.4308816024968367, + "learning_rate": 9.965342284774633e-06, + "loss": 0.704, + "step": 280 + }, + { + "epoch": 0.06679343950558593, + "grad_norm": 1.4937895092059466, + "learning_rate": 9.964888294928557e-06, + "loss": 0.717, + "step": 281 + }, + { + "epoch": 0.06703113857855954, + "grad_norm": 1.549773449919412, + "learning_rate": 9.964431361414122e-06, + "loss": 0.6791, + "step": 282 + }, + { + "epoch": 0.06726883765153316, + "grad_norm": 1.3235452812382726, + "learning_rate": 9.963971484502247e-06, + "loss": 0.6778, + "step": 283 + }, + { + "epoch": 0.06750653672450678, + "grad_norm": 1.1146517042634199, + "learning_rate": 9.963508664465586e-06, + "loss": 0.6756, + "step": 284 + }, + { + "epoch": 0.06774423579748039, + "grad_norm": 1.2303568244400853, + "learning_rate": 9.963042901578548e-06, + "loss": 0.7025, + "step": 285 + }, + { + "epoch": 0.06798193487045401, + "grad_norm": 1.8311952021934772, + "learning_rate": 9.962574196117281e-06, + "loss": 0.739, + "step": 286 + }, + { + "epoch": 0.06821963394342762, + "grad_norm": 1.3742409800536302, + "learning_rate": 9.96210254835968e-06, + "loss": 0.6736, + "step": 287 + }, + { + "epoch": 0.06845733301640124, + "grad_norm": 1.526937739914133, + "learning_rate": 9.961627958585382e-06, + "loss": 0.6443, + "step": 288 + }, + { + "epoch": 0.06869503208937486, + "grad_norm": 1.39649594671468, + "learning_rate": 9.96115042707577e-06, + "loss": 0.741, + "step": 289 + }, + { + "epoch": 0.06893273116234847, + "grad_norm": 1.2786204661296108, + "learning_rate": 9.960669954113974e-06, + "loss": 0.7294, + "step": 290 + }, + { + "epoch": 0.06917043023532209, + "grad_norm": 1.1950303072952217, + "learning_rate": 9.960186539984862e-06, + "loss": 0.6584, + "step": 291 + }, + { + "epoch": 0.0694081293082957, + "grad_norm": 1.3516502918321167, + "learning_rate": 9.95970018497505e-06, + "loss": 0.7374, + "step": 292 + }, + { + "epoch": 0.06964582838126931, + "grad_norm": 1.9796125426827234, + "learning_rate": 9.959210889372896e-06, + "loss": 0.6506, + "step": 293 + }, + { + "epoch": 0.06988352745424292, + "grad_norm": 1.3127696433014786, + "learning_rate": 9.958718653468501e-06, + "loss": 0.7407, + "step": 294 + }, + { + "epoch": 0.07012122652721654, + "grad_norm": 1.2727268195318375, + "learning_rate": 9.958223477553715e-06, + "loss": 0.6872, + "step": 295 + }, + { + "epoch": 0.07035892560019016, + "grad_norm": 1.4738044822967211, + "learning_rate": 9.957725361922121e-06, + "loss": 0.7023, + "step": 296 + }, + { + "epoch": 0.07059662467316377, + "grad_norm": 1.1937790201412504, + "learning_rate": 9.957224306869053e-06, + "loss": 0.6824, + "step": 297 + }, + { + "epoch": 0.07083432374613739, + "grad_norm": 1.2423844073403933, + "learning_rate": 9.956720312691584e-06, + "loss": 0.7042, + "step": 298 + }, + { + "epoch": 0.071072022819111, + "grad_norm": 1.1028157785871293, + "learning_rate": 9.956213379688531e-06, + "loss": 0.7056, + "step": 299 + }, + { + "epoch": 0.07130972189208462, + "grad_norm": 1.381003544980505, + "learning_rate": 9.955703508160454e-06, + "loss": 0.7147, + "step": 300 + }, + { + "epoch": 0.07130972189208462, + "eval_loss": 0.6954049468040466, + "eval_runtime": 277.5348, + "eval_samples_per_second": 39.195, + "eval_steps_per_second": 1.225, + "step": 300 + }, + { + "epoch": 0.07154742096505824, + "grad_norm": 1.3108935105391306, + "learning_rate": 9.955190698409656e-06, + "loss": 0.711, + "step": 301 + }, + { + "epoch": 0.07178512003803185, + "grad_norm": 1.2204767544962787, + "learning_rate": 9.954674950740175e-06, + "loss": 0.6922, + "step": 302 + }, + { + "epoch": 0.07202281911100547, + "grad_norm": 1.5498361427227028, + "learning_rate": 9.954156265457801e-06, + "loss": 0.6602, + "step": 303 + }, + { + "epoch": 0.07226051818397908, + "grad_norm": 1.523119851698443, + "learning_rate": 9.95363464287006e-06, + "loss": 0.6773, + "step": 304 + }, + { + "epoch": 0.0724982172569527, + "grad_norm": 1.3786794727804834, + "learning_rate": 9.95311008328622e-06, + "loss": 0.6989, + "step": 305 + }, + { + "epoch": 0.07273591632992632, + "grad_norm": 1.448053030986008, + "learning_rate": 9.952582587017293e-06, + "loss": 0.6856, + "step": 306 + }, + { + "epoch": 0.07297361540289993, + "grad_norm": 1.3675917851960724, + "learning_rate": 9.952052154376027e-06, + "loss": 0.6649, + "step": 307 + }, + { + "epoch": 0.07321131447587355, + "grad_norm": 1.385218902658616, + "learning_rate": 9.951518785676915e-06, + "loss": 0.6824, + "step": 308 + }, + { + "epoch": 0.07344901354884716, + "grad_norm": 1.4936667919625275, + "learning_rate": 9.95098248123619e-06, + "loss": 0.6223, + "step": 309 + }, + { + "epoch": 0.07368671262182078, + "grad_norm": 1.1059543573901736, + "learning_rate": 9.950443241371825e-06, + "loss": 0.631, + "step": 310 + }, + { + "epoch": 0.07392441169479438, + "grad_norm": 1.2540233563002905, + "learning_rate": 9.949901066403536e-06, + "loss": 0.6723, + "step": 311 + }, + { + "epoch": 0.074162110767768, + "grad_norm": 1.3622565957801895, + "learning_rate": 9.949355956652773e-06, + "loss": 0.7077, + "step": 312 + }, + { + "epoch": 0.07439980984074163, + "grad_norm": 1.3215928283424154, + "learning_rate": 9.948807912442735e-06, + "loss": 0.6855, + "step": 313 + }, + { + "epoch": 0.07463750891371523, + "grad_norm": 1.4378973150738261, + "learning_rate": 9.948256934098353e-06, + "loss": 0.6236, + "step": 314 + }, + { + "epoch": 0.07487520798668885, + "grad_norm": 1.3255378952296943, + "learning_rate": 9.9477030219463e-06, + "loss": 0.7078, + "step": 315 + }, + { + "epoch": 0.07511290705966246, + "grad_norm": 1.253168410092346, + "learning_rate": 9.947146176314988e-06, + "loss": 0.6741, + "step": 316 + }, + { + "epoch": 0.07535060613263608, + "grad_norm": 1.2285611353906032, + "learning_rate": 9.946586397534572e-06, + "loss": 0.6724, + "step": 317 + }, + { + "epoch": 0.0755883052056097, + "grad_norm": 1.321535930552684, + "learning_rate": 9.946023685936943e-06, + "loss": 0.7058, + "step": 318 + }, + { + "epoch": 0.07582600427858331, + "grad_norm": 1.27860010516669, + "learning_rate": 9.945458041855732e-06, + "loss": 0.6994, + "step": 319 + }, + { + "epoch": 0.07606370335155693, + "grad_norm": 1.2041668537718306, + "learning_rate": 9.944889465626303e-06, + "loss": 0.6743, + "step": 320 + }, + { + "epoch": 0.07630140242453054, + "grad_norm": 1.5146347557910695, + "learning_rate": 9.944317957585767e-06, + "loss": 0.7223, + "step": 321 + }, + { + "epoch": 0.07653910149750416, + "grad_norm": 1.3065130265132479, + "learning_rate": 9.943743518072971e-06, + "loss": 0.7502, + "step": 322 + }, + { + "epoch": 0.07677680057047777, + "grad_norm": 1.5270929094574959, + "learning_rate": 9.943166147428494e-06, + "loss": 0.6686, + "step": 323 + }, + { + "epoch": 0.07701449964345139, + "grad_norm": 1.1583734690016814, + "learning_rate": 9.94258584599466e-06, + "loss": 0.6368, + "step": 324 + }, + { + "epoch": 0.07725219871642501, + "grad_norm": 1.5247804999824237, + "learning_rate": 9.94200261411553e-06, + "loss": 0.677, + "step": 325 + }, + { + "epoch": 0.07748989778939862, + "grad_norm": 1.2757415696098868, + "learning_rate": 9.941416452136899e-06, + "loss": 0.7262, + "step": 326 + }, + { + "epoch": 0.07772759686237224, + "grad_norm": 1.2229897751870007, + "learning_rate": 9.940827360406297e-06, + "loss": 0.6713, + "step": 327 + }, + { + "epoch": 0.07796529593534585, + "grad_norm": 1.3157100650884153, + "learning_rate": 9.940235339273002e-06, + "loss": 0.7056, + "step": 328 + }, + { + "epoch": 0.07820299500831947, + "grad_norm": 1.94129401504843, + "learning_rate": 9.939640389088015e-06, + "loss": 0.6917, + "step": 329 + }, + { + "epoch": 0.07844069408129309, + "grad_norm": 1.4614645157194481, + "learning_rate": 9.939042510204084e-06, + "loss": 0.7104, + "step": 330 + }, + { + "epoch": 0.0786783931542667, + "grad_norm": 1.1868022028031304, + "learning_rate": 9.938441702975689e-06, + "loss": 0.6768, + "step": 331 + }, + { + "epoch": 0.07891609222724032, + "grad_norm": 1.1976748722272643, + "learning_rate": 9.937837967759047e-06, + "loss": 0.6847, + "step": 332 + }, + { + "epoch": 0.07915379130021392, + "grad_norm": 1.195824396979555, + "learning_rate": 9.937231304912112e-06, + "loss": 0.6831, + "step": 333 + }, + { + "epoch": 0.07939149037318755, + "grad_norm": 1.2875026397802218, + "learning_rate": 9.93662171479457e-06, + "loss": 0.692, + "step": 334 + }, + { + "epoch": 0.07962918944616117, + "grad_norm": 1.3344057878467104, + "learning_rate": 9.936009197767847e-06, + "loss": 0.6972, + "step": 335 + }, + { + "epoch": 0.07986688851913477, + "grad_norm": 1.3323184552080418, + "learning_rate": 9.935393754195101e-06, + "loss": 0.7036, + "step": 336 + }, + { + "epoch": 0.0801045875921084, + "grad_norm": 1.4721724102441063, + "learning_rate": 9.93477538444123e-06, + "loss": 0.7037, + "step": 337 + }, + { + "epoch": 0.080342286665082, + "grad_norm": 2.0425122890559724, + "learning_rate": 9.93415408887286e-06, + "loss": 0.654, + "step": 338 + }, + { + "epoch": 0.08057998573805562, + "grad_norm": 1.9297976544059845, + "learning_rate": 9.933529867858358e-06, + "loss": 0.6893, + "step": 339 + }, + { + "epoch": 0.08081768481102923, + "grad_norm": 1.838002056459463, + "learning_rate": 9.932902721767823e-06, + "loss": 0.6505, + "step": 340 + }, + { + "epoch": 0.08105538388400285, + "grad_norm": 1.1641136216815153, + "learning_rate": 9.932272650973085e-06, + "loss": 0.6757, + "step": 341 + }, + { + "epoch": 0.08129308295697647, + "grad_norm": 1.3463653146779295, + "learning_rate": 9.931639655847716e-06, + "loss": 0.6898, + "step": 342 + }, + { + "epoch": 0.08153078202995008, + "grad_norm": 1.2970749985690126, + "learning_rate": 9.931003736767013e-06, + "loss": 0.6908, + "step": 343 + }, + { + "epoch": 0.0817684811029237, + "grad_norm": 1.5536271147906247, + "learning_rate": 9.930364894108013e-06, + "loss": 0.671, + "step": 344 + }, + { + "epoch": 0.08200618017589731, + "grad_norm": 1.2051268673917102, + "learning_rate": 9.929723128249483e-06, + "loss": 0.6736, + "step": 345 + }, + { + "epoch": 0.08224387924887093, + "grad_norm": 1.3094269256855606, + "learning_rate": 9.929078439571925e-06, + "loss": 0.6434, + "step": 346 + }, + { + "epoch": 0.08248157832184455, + "grad_norm": 1.4333862437243592, + "learning_rate": 9.92843082845757e-06, + "loss": 0.6925, + "step": 347 + }, + { + "epoch": 0.08271927739481816, + "grad_norm": 1.2325285137528483, + "learning_rate": 9.92778029529039e-06, + "loss": 0.6914, + "step": 348 + }, + { + "epoch": 0.08295697646779178, + "grad_norm": 1.2712065425082113, + "learning_rate": 9.92712684045608e-06, + "loss": 0.6928, + "step": 349 + }, + { + "epoch": 0.08319467554076539, + "grad_norm": 1.159067899510439, + "learning_rate": 9.926470464342073e-06, + "loss": 0.6776, + "step": 350 + }, + { + "epoch": 0.08343237461373901, + "grad_norm": 1.1920858592624373, + "learning_rate": 9.925811167337533e-06, + "loss": 0.7402, + "step": 351 + }, + { + "epoch": 0.08367007368671261, + "grad_norm": 1.222616624735708, + "learning_rate": 9.925148949833356e-06, + "loss": 0.6956, + "step": 352 + }, + { + "epoch": 0.08390777275968624, + "grad_norm": 1.2051411147044915, + "learning_rate": 9.924483812222166e-06, + "loss": 0.6439, + "step": 353 + }, + { + "epoch": 0.08414547183265986, + "grad_norm": 1.1791176993963517, + "learning_rate": 9.923815754898325e-06, + "loss": 0.7481, + "step": 354 + }, + { + "epoch": 0.08438317090563346, + "grad_norm": 1.2723430295437272, + "learning_rate": 9.923144778257918e-06, + "loss": 0.6947, + "step": 355 + }, + { + "epoch": 0.08462086997860709, + "grad_norm": 1.1965796982176629, + "learning_rate": 9.92247088269877e-06, + "loss": 0.6674, + "step": 356 + }, + { + "epoch": 0.08485856905158069, + "grad_norm": 1.30461992266437, + "learning_rate": 9.92179406862043e-06, + "loss": 0.6979, + "step": 357 + }, + { + "epoch": 0.08509626812455431, + "grad_norm": 1.1672827966300077, + "learning_rate": 9.921114336424176e-06, + "loss": 0.697, + "step": 358 + }, + { + "epoch": 0.08533396719752794, + "grad_norm": 1.3426998650895425, + "learning_rate": 9.920431686513023e-06, + "loss": 0.7017, + "step": 359 + }, + { + "epoch": 0.08557166627050154, + "grad_norm": 1.1836210097863726, + "learning_rate": 9.91974611929171e-06, + "loss": 0.6975, + "step": 360 + }, + { + "epoch": 0.08580936534347516, + "grad_norm": 1.3685599260952932, + "learning_rate": 9.91905763516671e-06, + "loss": 0.6962, + "step": 361 + }, + { + "epoch": 0.08604706441644877, + "grad_norm": 1.2597405759935687, + "learning_rate": 9.918366234546223e-06, + "loss": 0.7158, + "step": 362 + }, + { + "epoch": 0.08628476348942239, + "grad_norm": 1.253326768483047, + "learning_rate": 9.917671917840178e-06, + "loss": 0.7053, + "step": 363 + }, + { + "epoch": 0.08652246256239601, + "grad_norm": 1.2310175269645576, + "learning_rate": 9.916974685460232e-06, + "loss": 0.6965, + "step": 364 + }, + { + "epoch": 0.08676016163536962, + "grad_norm": 1.1442920740610785, + "learning_rate": 9.916274537819774e-06, + "loss": 0.6904, + "step": 365 + }, + { + "epoch": 0.08699786070834324, + "grad_norm": 1.3238029377631597, + "learning_rate": 9.915571475333919e-06, + "loss": 0.6768, + "step": 366 + }, + { + "epoch": 0.08723555978131685, + "grad_norm": 1.2718132485196483, + "learning_rate": 9.91486549841951e-06, + "loss": 0.7223, + "step": 367 + }, + { + "epoch": 0.08747325885429047, + "grad_norm": 1.281638856499706, + "learning_rate": 9.91415660749512e-06, + "loss": 0.6348, + "step": 368 + }, + { + "epoch": 0.08771095792726408, + "grad_norm": 1.4391199562551062, + "learning_rate": 9.913444802981047e-06, + "loss": 0.7054, + "step": 369 + }, + { + "epoch": 0.0879486570002377, + "grad_norm": 6.330249159353639, + "learning_rate": 9.912730085299321e-06, + "loss": 0.6673, + "step": 370 + }, + { + "epoch": 0.08818635607321132, + "grad_norm": 1.166045375779884, + "learning_rate": 9.912012454873691e-06, + "loss": 0.7044, + "step": 371 + }, + { + "epoch": 0.08842405514618493, + "grad_norm": 1.8416868733414489, + "learning_rate": 9.911291912129642e-06, + "loss": 0.7214, + "step": 372 + }, + { + "epoch": 0.08866175421915855, + "grad_norm": 1.1890704799504244, + "learning_rate": 9.91056845749438e-06, + "loss": 0.6346, + "step": 373 + }, + { + "epoch": 0.08889945329213216, + "grad_norm": 1.2430956201692185, + "learning_rate": 9.909842091396842e-06, + "loss": 0.6971, + "step": 374 + }, + { + "epoch": 0.08913715236510578, + "grad_norm": 1.4232150459821826, + "learning_rate": 9.909112814267686e-06, + "loss": 0.6601, + "step": 375 + }, + { + "epoch": 0.0893748514380794, + "grad_norm": 1.1956288742491998, + "learning_rate": 9.908380626539297e-06, + "loss": 0.6754, + "step": 376 + }, + { + "epoch": 0.089612550511053, + "grad_norm": 1.21077571104721, + "learning_rate": 9.907645528645791e-06, + "loss": 0.7191, + "step": 377 + }, + { + "epoch": 0.08985024958402663, + "grad_norm": 1.4928388796676335, + "learning_rate": 9.906907521023e-06, + "loss": 0.6378, + "step": 378 + }, + { + "epoch": 0.09008794865700023, + "grad_norm": 1.415636614795306, + "learning_rate": 9.906166604108494e-06, + "loss": 0.6937, + "step": 379 + }, + { + "epoch": 0.09032564772997385, + "grad_norm": 1.3370859944440026, + "learning_rate": 9.905422778341554e-06, + "loss": 0.706, + "step": 380 + }, + { + "epoch": 0.09056334680294748, + "grad_norm": 1.8183377490004586, + "learning_rate": 9.904676044163195e-06, + "loss": 0.7216, + "step": 381 + }, + { + "epoch": 0.09080104587592108, + "grad_norm": 1.2558146483573438, + "learning_rate": 9.903926402016153e-06, + "loss": 0.6912, + "step": 382 + }, + { + "epoch": 0.0910387449488947, + "grad_norm": 1.2155727731964228, + "learning_rate": 9.903173852344889e-06, + "loss": 0.6799, + "step": 383 + }, + { + "epoch": 0.09127644402186831, + "grad_norm": 1.2832601673705042, + "learning_rate": 9.902418395595589e-06, + "loss": 0.6935, + "step": 384 + }, + { + "epoch": 0.09151414309484193, + "grad_norm": 1.1844681637648153, + "learning_rate": 9.901660032216159e-06, + "loss": 0.6656, + "step": 385 + }, + { + "epoch": 0.09175184216781554, + "grad_norm": 1.42758006746676, + "learning_rate": 9.900898762656231e-06, + "loss": 0.726, + "step": 386 + }, + { + "epoch": 0.09198954124078916, + "grad_norm": 1.6794579832806837, + "learning_rate": 9.90013458736716e-06, + "loss": 0.6908, + "step": 387 + }, + { + "epoch": 0.09222724031376278, + "grad_norm": 1.4882605719548216, + "learning_rate": 9.899367506802023e-06, + "loss": 0.684, + "step": 388 + }, + { + "epoch": 0.09246493938673639, + "grad_norm": 1.4971436507177236, + "learning_rate": 9.898597521415622e-06, + "loss": 0.6965, + "step": 389 + }, + { + "epoch": 0.09270263845971001, + "grad_norm": 1.3841591015431594, + "learning_rate": 9.897824631664476e-06, + "loss": 0.6468, + "step": 390 + }, + { + "epoch": 0.09294033753268362, + "grad_norm": 1.66312628203809, + "learning_rate": 9.89704883800683e-06, + "loss": 0.6828, + "step": 391 + }, + { + "epoch": 0.09317803660565724, + "grad_norm": 1.218213171926895, + "learning_rate": 9.89627014090265e-06, + "loss": 0.7241, + "step": 392 + }, + { + "epoch": 0.09341573567863086, + "grad_norm": 1.2297710462907654, + "learning_rate": 9.895488540813628e-06, + "loss": 0.6521, + "step": 393 + }, + { + "epoch": 0.09365343475160447, + "grad_norm": 1.2614869433756346, + "learning_rate": 9.894704038203164e-06, + "loss": 0.7041, + "step": 394 + }, + { + "epoch": 0.09389113382457809, + "grad_norm": 1.2404290593688945, + "learning_rate": 9.893916633536393e-06, + "loss": 0.7252, + "step": 395 + }, + { + "epoch": 0.0941288328975517, + "grad_norm": 1.4349871636038851, + "learning_rate": 9.893126327280163e-06, + "loss": 0.6398, + "step": 396 + }, + { + "epoch": 0.09436653197052532, + "grad_norm": 1.1937403903864428, + "learning_rate": 9.892333119903045e-06, + "loss": 0.7443, + "step": 397 + }, + { + "epoch": 0.09460423104349892, + "grad_norm": 1.1091888705839994, + "learning_rate": 9.891537011875332e-06, + "loss": 0.6521, + "step": 398 + }, + { + "epoch": 0.09484193011647254, + "grad_norm": 1.1784773318944892, + "learning_rate": 9.890738003669029e-06, + "loss": 0.7145, + "step": 399 + }, + { + "epoch": 0.09507962918944617, + "grad_norm": 1.305505907103123, + "learning_rate": 9.88993609575787e-06, + "loss": 0.6687, + "step": 400 + }, + { + "epoch": 0.09507962918944617, + "eval_loss": 0.6863201856613159, + "eval_runtime": 277.8542, + "eval_samples_per_second": 39.15, + "eval_steps_per_second": 1.224, + "step": 400 + }, + { + "epoch": 0.09531732826241977, + "grad_norm": 1.3680812996246503, + "learning_rate": 9.889131288617301e-06, + "loss": 0.6971, + "step": 401 + }, + { + "epoch": 0.0955550273353934, + "grad_norm": 1.1948250203967763, + "learning_rate": 9.888323582724493e-06, + "loss": 0.6998, + "step": 402 + }, + { + "epoch": 0.095792726408367, + "grad_norm": 1.1547954188467981, + "learning_rate": 9.887512978558329e-06, + "loss": 0.6955, + "step": 403 + }, + { + "epoch": 0.09603042548134062, + "grad_norm": 1.3379931466449149, + "learning_rate": 9.886699476599417e-06, + "loss": 0.7066, + "step": 404 + }, + { + "epoch": 0.09626812455431424, + "grad_norm": 1.2591120143678791, + "learning_rate": 9.885883077330078e-06, + "loss": 0.696, + "step": 405 + }, + { + "epoch": 0.09650582362728785, + "grad_norm": 1.3095214642241864, + "learning_rate": 9.885063781234356e-06, + "loss": 0.6937, + "step": 406 + }, + { + "epoch": 0.09674352270026147, + "grad_norm": 1.145881368543689, + "learning_rate": 9.884241588798004e-06, + "loss": 0.7178, + "step": 407 + }, + { + "epoch": 0.09698122177323508, + "grad_norm": 1.2956436166360008, + "learning_rate": 9.883416500508502e-06, + "loss": 0.7164, + "step": 408 + }, + { + "epoch": 0.0972189208462087, + "grad_norm": 1.2713002362085082, + "learning_rate": 9.88258851685504e-06, + "loss": 0.6582, + "step": 409 + }, + { + "epoch": 0.09745661991918232, + "grad_norm": 1.263182994327568, + "learning_rate": 9.88175763832853e-06, + "loss": 0.6524, + "step": 410 + }, + { + "epoch": 0.09769431899215593, + "grad_norm": 2.419598865116206, + "learning_rate": 9.880923865421597e-06, + "loss": 0.6735, + "step": 411 + }, + { + "epoch": 0.09793201806512955, + "grad_norm": 1.2198129993503517, + "learning_rate": 9.880087198628579e-06, + "loss": 0.6405, + "step": 412 + }, + { + "epoch": 0.09816971713810316, + "grad_norm": 1.4011149439594726, + "learning_rate": 9.879247638445536e-06, + "loss": 0.75, + "step": 413 + }, + { + "epoch": 0.09840741621107678, + "grad_norm": 1.7822175592991607, + "learning_rate": 9.878405185370244e-06, + "loss": 0.6713, + "step": 414 + }, + { + "epoch": 0.09864511528405039, + "grad_norm": 1.459195435184289, + "learning_rate": 9.877559839902185e-06, + "loss": 0.6914, + "step": 415 + }, + { + "epoch": 0.09888281435702401, + "grad_norm": 1.3519055572325842, + "learning_rate": 9.876711602542564e-06, + "loss": 0.7228, + "step": 416 + }, + { + "epoch": 0.09912051342999763, + "grad_norm": 1.0687612110111064, + "learning_rate": 9.875860473794302e-06, + "loss": 0.6953, + "step": 417 + }, + { + "epoch": 0.09935821250297124, + "grad_norm": 1.2773168710237308, + "learning_rate": 9.875006454162027e-06, + "loss": 0.7051, + "step": 418 + }, + { + "epoch": 0.09959591157594486, + "grad_norm": 1.8830228788866914, + "learning_rate": 9.874149544152085e-06, + "loss": 0.7132, + "step": 419 + }, + { + "epoch": 0.09983361064891846, + "grad_norm": 1.2716396046973777, + "learning_rate": 9.873289744272537e-06, + "loss": 0.7363, + "step": 420 + }, + { + "epoch": 0.10007130972189208, + "grad_norm": 1.388022247636439, + "learning_rate": 9.872427055033156e-06, + "loss": 0.6791, + "step": 421 + }, + { + "epoch": 0.1003090087948657, + "grad_norm": 1.1821263068670442, + "learning_rate": 9.871561476945429e-06, + "loss": 0.7093, + "step": 422 + }, + { + "epoch": 0.10054670786783931, + "grad_norm": 1.1520454636338981, + "learning_rate": 9.870693010522552e-06, + "loss": 0.6826, + "step": 423 + }, + { + "epoch": 0.10078440694081293, + "grad_norm": 1.2778543131960929, + "learning_rate": 9.869821656279438e-06, + "loss": 0.6947, + "step": 424 + }, + { + "epoch": 0.10102210601378654, + "grad_norm": 1.287888456391539, + "learning_rate": 9.86894741473271e-06, + "loss": 0.6724, + "step": 425 + }, + { + "epoch": 0.10125980508676016, + "grad_norm": 1.1932968870365175, + "learning_rate": 9.868070286400702e-06, + "loss": 0.6617, + "step": 426 + }, + { + "epoch": 0.10149750415973377, + "grad_norm": 1.1873386308078682, + "learning_rate": 9.867190271803466e-06, + "loss": 0.6492, + "step": 427 + }, + { + "epoch": 0.10173520323270739, + "grad_norm": 1.1324600262263984, + "learning_rate": 9.866307371462753e-06, + "loss": 0.6621, + "step": 428 + }, + { + "epoch": 0.10197290230568101, + "grad_norm": 1.5112802893436046, + "learning_rate": 9.865421585902039e-06, + "loss": 0.6686, + "step": 429 + }, + { + "epoch": 0.10221060137865462, + "grad_norm": 1.682324054506429, + "learning_rate": 9.864532915646498e-06, + "loss": 0.7472, + "step": 430 + }, + { + "epoch": 0.10244830045162824, + "grad_norm": 1.5338697160454893, + "learning_rate": 9.863641361223025e-06, + "loss": 0.6938, + "step": 431 + }, + { + "epoch": 0.10268599952460185, + "grad_norm": 1.174708770759674, + "learning_rate": 9.862746923160215e-06, + "loss": 0.6917, + "step": 432 + }, + { + "epoch": 0.10292369859757547, + "grad_norm": 1.3018488724939299, + "learning_rate": 9.861849601988384e-06, + "loss": 0.6875, + "step": 433 + }, + { + "epoch": 0.10316139767054909, + "grad_norm": 1.2731433472076905, + "learning_rate": 9.860949398239545e-06, + "loss": 0.6773, + "step": 434 + }, + { + "epoch": 0.1033990967435227, + "grad_norm": 1.309962892581243, + "learning_rate": 9.860046312447432e-06, + "loss": 0.6636, + "step": 435 + }, + { + "epoch": 0.10363679581649632, + "grad_norm": 1.0578788028499715, + "learning_rate": 9.859140345147477e-06, + "loss": 0.6777, + "step": 436 + }, + { + "epoch": 0.10387449488946993, + "grad_norm": 1.1203292750470564, + "learning_rate": 9.85823149687683e-06, + "loss": 0.7029, + "step": 437 + }, + { + "epoch": 0.10411219396244355, + "grad_norm": 1.2836246699647307, + "learning_rate": 9.857319768174339e-06, + "loss": 0.6474, + "step": 438 + }, + { + "epoch": 0.10434989303541717, + "grad_norm": 1.5733073604723828, + "learning_rate": 9.85640515958057e-06, + "loss": 0.6761, + "step": 439 + }, + { + "epoch": 0.10458759210839078, + "grad_norm": 1.5639674866873123, + "learning_rate": 9.855487671637791e-06, + "loss": 0.7057, + "step": 440 + }, + { + "epoch": 0.1048252911813644, + "grad_norm": 1.6673443369447154, + "learning_rate": 9.854567304889977e-06, + "loss": 0.6885, + "step": 441 + }, + { + "epoch": 0.105062990254338, + "grad_norm": 1.1804765114050977, + "learning_rate": 9.853644059882812e-06, + "loss": 0.6872, + "step": 442 + }, + { + "epoch": 0.10530068932731163, + "grad_norm": 1.5545295677574364, + "learning_rate": 9.852717937163683e-06, + "loss": 0.6316, + "step": 443 + }, + { + "epoch": 0.10553838840028523, + "grad_norm": 1.2402471436592437, + "learning_rate": 9.85178893728169e-06, + "loss": 0.6724, + "step": 444 + }, + { + "epoch": 0.10577608747325885, + "grad_norm": 1.4533123531887882, + "learning_rate": 9.85085706078763e-06, + "loss": 0.6967, + "step": 445 + }, + { + "epoch": 0.10601378654623247, + "grad_norm": 1.3495478050513452, + "learning_rate": 9.849922308234013e-06, + "loss": 0.713, + "step": 446 + }, + { + "epoch": 0.10625148561920608, + "grad_norm": 1.123853957328718, + "learning_rate": 9.848984680175049e-06, + "loss": 0.6674, + "step": 447 + }, + { + "epoch": 0.1064891846921797, + "grad_norm": 1.6294993483641256, + "learning_rate": 9.848044177166656e-06, + "loss": 0.6379, + "step": 448 + }, + { + "epoch": 0.10672688376515331, + "grad_norm": 1.3455474736344117, + "learning_rate": 9.847100799766455e-06, + "loss": 0.6786, + "step": 449 + }, + { + "epoch": 0.10696458283812693, + "grad_norm": 1.2994763841189825, + "learning_rate": 9.846154548533773e-06, + "loss": 0.6606, + "step": 450 + }, + { + "epoch": 0.10720228191110055, + "grad_norm": 1.23698133708254, + "learning_rate": 9.845205424029639e-06, + "loss": 0.6957, + "step": 451 + }, + { + "epoch": 0.10743998098407416, + "grad_norm": 1.2656463898076062, + "learning_rate": 9.844253426816785e-06, + "loss": 0.655, + "step": 452 + }, + { + "epoch": 0.10767768005704778, + "grad_norm": 1.2251647188589372, + "learning_rate": 9.843298557459651e-06, + "loss": 0.6332, + "step": 453 + }, + { + "epoch": 0.10791537913002139, + "grad_norm": 2.0869239680422207, + "learning_rate": 9.842340816524374e-06, + "loss": 0.6898, + "step": 454 + }, + { + "epoch": 0.10815307820299501, + "grad_norm": 1.3351855634018548, + "learning_rate": 9.841380204578795e-06, + "loss": 0.6227, + "step": 455 + }, + { + "epoch": 0.10839077727596863, + "grad_norm": 1.341618414281126, + "learning_rate": 9.84041672219246e-06, + "loss": 0.6492, + "step": 456 + }, + { + "epoch": 0.10862847634894224, + "grad_norm": 1.2980428600832423, + "learning_rate": 9.839450369936615e-06, + "loss": 0.7041, + "step": 457 + }, + { + "epoch": 0.10886617542191586, + "grad_norm": 1.164567698098084, + "learning_rate": 9.838481148384205e-06, + "loss": 0.6804, + "step": 458 + }, + { + "epoch": 0.10910387449488947, + "grad_norm": 1.354601361605155, + "learning_rate": 9.83750905810988e-06, + "loss": 0.6995, + "step": 459 + }, + { + "epoch": 0.10934157356786309, + "grad_norm": 1.237707756001022, + "learning_rate": 9.836534099689992e-06, + "loss": 0.7024, + "step": 460 + }, + { + "epoch": 0.1095792726408367, + "grad_norm": 1.127362620964333, + "learning_rate": 9.835556273702589e-06, + "loss": 0.6453, + "step": 461 + }, + { + "epoch": 0.10981697171381032, + "grad_norm": 1.118239356907589, + "learning_rate": 9.83457558072742e-06, + "loss": 0.6194, + "step": 462 + }, + { + "epoch": 0.11005467078678394, + "grad_norm": 1.2398168944198655, + "learning_rate": 9.833592021345938e-06, + "loss": 0.7185, + "step": 463 + }, + { + "epoch": 0.11029236985975754, + "grad_norm": 1.3188245598902986, + "learning_rate": 9.832605596141292e-06, + "loss": 0.6541, + "step": 464 + }, + { + "epoch": 0.11053006893273117, + "grad_norm": 1.1278749414568585, + "learning_rate": 9.831616305698329e-06, + "loss": 0.6342, + "step": 465 + }, + { + "epoch": 0.11076776800570477, + "grad_norm": 1.308784621302563, + "learning_rate": 9.830624150603596e-06, + "loss": 0.6887, + "step": 466 + }, + { + "epoch": 0.1110054670786784, + "grad_norm": 1.6954097093207032, + "learning_rate": 9.829629131445342e-06, + "loss": 0.7119, + "step": 467 + }, + { + "epoch": 0.11124316615165201, + "grad_norm": 2.2260234363261873, + "learning_rate": 9.82863124881351e-06, + "loss": 0.6857, + "step": 468 + }, + { + "epoch": 0.11148086522462562, + "grad_norm": 1.417286336597015, + "learning_rate": 9.827630503299741e-06, + "loss": 0.6981, + "step": 469 + }, + { + "epoch": 0.11171856429759924, + "grad_norm": 1.2721381264711369, + "learning_rate": 9.826626895497373e-06, + "loss": 0.6117, + "step": 470 + }, + { + "epoch": 0.11195626337057285, + "grad_norm": 1.230343229687748, + "learning_rate": 9.825620426001446e-06, + "loss": 0.7009, + "step": 471 + }, + { + "epoch": 0.11219396244354647, + "grad_norm": 1.3465717270858404, + "learning_rate": 9.824611095408691e-06, + "loss": 0.6753, + "step": 472 + }, + { + "epoch": 0.11243166151652008, + "grad_norm": 1.4289998363736924, + "learning_rate": 9.823598904317535e-06, + "loss": 0.7025, + "step": 473 + }, + { + "epoch": 0.1126693605894937, + "grad_norm": 1.3513012523187824, + "learning_rate": 9.822583853328106e-06, + "loss": 0.6657, + "step": 474 + }, + { + "epoch": 0.11290705966246732, + "grad_norm": 2.297980692445176, + "learning_rate": 9.821565943042225e-06, + "loss": 0.6625, + "step": 475 + }, + { + "epoch": 0.11314475873544093, + "grad_norm": 1.1581893651479325, + "learning_rate": 9.820545174063407e-06, + "loss": 0.6624, + "step": 476 + }, + { + "epoch": 0.11338245780841455, + "grad_norm": 1.2297438110121346, + "learning_rate": 9.819521546996864e-06, + "loss": 0.6697, + "step": 477 + }, + { + "epoch": 0.11362015688138816, + "grad_norm": 1.185241071146067, + "learning_rate": 9.8184950624495e-06, + "loss": 0.6337, + "step": 478 + }, + { + "epoch": 0.11385785595436178, + "grad_norm": 1.1936596721503125, + "learning_rate": 9.817465721029916e-06, + "loss": 0.7052, + "step": 479 + }, + { + "epoch": 0.1140955550273354, + "grad_norm": 1.3777805232011804, + "learning_rate": 9.816433523348407e-06, + "loss": 0.6619, + "step": 480 + }, + { + "epoch": 0.114333254100309, + "grad_norm": 1.2114019590483547, + "learning_rate": 9.815398470016957e-06, + "loss": 0.6809, + "step": 481 + }, + { + "epoch": 0.11457095317328263, + "grad_norm": 1.4702308366141927, + "learning_rate": 9.814360561649247e-06, + "loss": 0.734, + "step": 482 + }, + { + "epoch": 0.11480865224625623, + "grad_norm": 1.139856847017618, + "learning_rate": 9.813319798860654e-06, + "loss": 0.6956, + "step": 483 + }, + { + "epoch": 0.11504635131922986, + "grad_norm": 1.5221886915639382, + "learning_rate": 9.812276182268236e-06, + "loss": 0.6916, + "step": 484 + }, + { + "epoch": 0.11528405039220348, + "grad_norm": 1.2604412655723771, + "learning_rate": 9.811229712490757e-06, + "loss": 0.6598, + "step": 485 + }, + { + "epoch": 0.11552174946517708, + "grad_norm": 5.198183207577298, + "learning_rate": 9.810180390148664e-06, + "loss": 0.764, + "step": 486 + }, + { + "epoch": 0.1157594485381507, + "grad_norm": 1.4490626843413885, + "learning_rate": 9.809128215864096e-06, + "loss": 0.6189, + "step": 487 + }, + { + "epoch": 0.11599714761112431, + "grad_norm": 1.4059662763236664, + "learning_rate": 9.808073190260886e-06, + "loss": 0.6747, + "step": 488 + }, + { + "epoch": 0.11623484668409793, + "grad_norm": 1.3619359678190524, + "learning_rate": 9.807015313964553e-06, + "loss": 0.6929, + "step": 489 + }, + { + "epoch": 0.11647254575707154, + "grad_norm": 1.2398820947361109, + "learning_rate": 9.80595458760231e-06, + "loss": 0.6677, + "step": 490 + }, + { + "epoch": 0.11671024483004516, + "grad_norm": 1.2098590472368014, + "learning_rate": 9.804891011803062e-06, + "loss": 0.681, + "step": 491 + }, + { + "epoch": 0.11694794390301878, + "grad_norm": 1.3756949333363908, + "learning_rate": 9.803824587197395e-06, + "loss": 0.7491, + "step": 492 + }, + { + "epoch": 0.11718564297599239, + "grad_norm": 1.1803368064459672, + "learning_rate": 9.802755314417592e-06, + "loss": 0.6767, + "step": 493 + }, + { + "epoch": 0.11742334204896601, + "grad_norm": 1.5331334817411932, + "learning_rate": 9.801683194097622e-06, + "loss": 0.6705, + "step": 494 + }, + { + "epoch": 0.11766104112193962, + "grad_norm": 1.2940273964973699, + "learning_rate": 9.800608226873143e-06, + "loss": 0.6353, + "step": 495 + }, + { + "epoch": 0.11789874019491324, + "grad_norm": 1.3549174889113016, + "learning_rate": 9.799530413381494e-06, + "loss": 0.689, + "step": 496 + }, + { + "epoch": 0.11813643926788686, + "grad_norm": 1.1857366934462301, + "learning_rate": 9.798449754261716e-06, + "loss": 0.6405, + "step": 497 + }, + { + "epoch": 0.11837413834086047, + "grad_norm": 1.3168060907145038, + "learning_rate": 9.797366250154524e-06, + "loss": 0.611, + "step": 498 + }, + { + "epoch": 0.11861183741383409, + "grad_norm": 1.448676047823134, + "learning_rate": 9.796279901702326e-06, + "loss": 0.6817, + "step": 499 + }, + { + "epoch": 0.1188495364868077, + "grad_norm": 1.361332723905112, + "learning_rate": 9.795190709549214e-06, + "loss": 0.64, + "step": 500 + }, + { + "epoch": 0.1188495364868077, + "eval_loss": 0.6777105927467346, + "eval_runtime": 284.4206, + "eval_samples_per_second": 38.246, + "eval_steps_per_second": 1.195, + "step": 500 + }, + { + "epoch": 0.11908723555978132, + "grad_norm": 1.2650986901512493, + "learning_rate": 9.794098674340966e-06, + "loss": 0.6591, + "step": 501 + }, + { + "epoch": 0.11932493463275493, + "grad_norm": 1.5057664831158044, + "learning_rate": 9.793003796725049e-06, + "loss": 0.6578, + "step": 502 + }, + { + "epoch": 0.11956263370572855, + "grad_norm": 1.2165719654267249, + "learning_rate": 9.791906077350613e-06, + "loss": 0.7171, + "step": 503 + }, + { + "epoch": 0.11980033277870217, + "grad_norm": 1.5481088819670192, + "learning_rate": 9.79080551686849e-06, + "loss": 0.6692, + "step": 504 + }, + { + "epoch": 0.12003803185167577, + "grad_norm": 1.4094934766842775, + "learning_rate": 9.789702115931202e-06, + "loss": 0.73, + "step": 505 + }, + { + "epoch": 0.1202757309246494, + "grad_norm": 1.242910098607317, + "learning_rate": 9.78859587519295e-06, + "loss": 0.6717, + "step": 506 + }, + { + "epoch": 0.120513429997623, + "grad_norm": 1.4602649507214698, + "learning_rate": 9.787486795309621e-06, + "loss": 0.6424, + "step": 507 + }, + { + "epoch": 0.12075112907059662, + "grad_norm": 1.291233802055334, + "learning_rate": 9.78637487693879e-06, + "loss": 0.7057, + "step": 508 + }, + { + "epoch": 0.12098882814357025, + "grad_norm": 1.391554588594508, + "learning_rate": 9.785260120739707e-06, + "loss": 0.6565, + "step": 509 + }, + { + "epoch": 0.12122652721654385, + "grad_norm": 1.166557161136071, + "learning_rate": 9.784142527373304e-06, + "loss": 0.6665, + "step": 510 + }, + { + "epoch": 0.12146422628951747, + "grad_norm": 1.3165448484936935, + "learning_rate": 9.783022097502204e-06, + "loss": 0.6614, + "step": 511 + }, + { + "epoch": 0.12170192536249108, + "grad_norm": 1.7862825313258746, + "learning_rate": 9.781898831790706e-06, + "loss": 0.6361, + "step": 512 + }, + { + "epoch": 0.1219396244354647, + "grad_norm": 1.2283672702760633, + "learning_rate": 9.78077273090479e-06, + "loss": 0.6767, + "step": 513 + }, + { + "epoch": 0.12217732350843832, + "grad_norm": 1.3784775785457348, + "learning_rate": 9.779643795512119e-06, + "loss": 0.65, + "step": 514 + }, + { + "epoch": 0.12241502258141193, + "grad_norm": 1.3048550495800104, + "learning_rate": 9.778512026282037e-06, + "loss": 0.6849, + "step": 515 + }, + { + "epoch": 0.12265272165438555, + "grad_norm": 1.4433368171927354, + "learning_rate": 9.777377423885562e-06, + "loss": 0.6703, + "step": 516 + }, + { + "epoch": 0.12289042072735916, + "grad_norm": 1.1301118796155372, + "learning_rate": 9.776239988995401e-06, + "loss": 0.6365, + "step": 517 + }, + { + "epoch": 0.12312811980033278, + "grad_norm": 1.2871923089481143, + "learning_rate": 9.775099722285934e-06, + "loss": 0.6165, + "step": 518 + }, + { + "epoch": 0.12336581887330639, + "grad_norm": 1.6451397190575896, + "learning_rate": 9.773956624433224e-06, + "loss": 0.7062, + "step": 519 + }, + { + "epoch": 0.12360351794628001, + "grad_norm": 1.1664794194719696, + "learning_rate": 9.772810696115006e-06, + "loss": 0.6348, + "step": 520 + }, + { + "epoch": 0.12384121701925363, + "grad_norm": 1.1256542737083013, + "learning_rate": 9.771661938010701e-06, + "loss": 0.6379, + "step": 521 + }, + { + "epoch": 0.12407891609222724, + "grad_norm": 1.1743101529781967, + "learning_rate": 9.770510350801407e-06, + "loss": 0.6881, + "step": 522 + }, + { + "epoch": 0.12431661516520086, + "grad_norm": 1.1955237774522811, + "learning_rate": 9.76935593516989e-06, + "loss": 0.6387, + "step": 523 + }, + { + "epoch": 0.12455431423817447, + "grad_norm": 1.1988165365116747, + "learning_rate": 9.768198691800607e-06, + "loss": 0.6343, + "step": 524 + }, + { + "epoch": 0.12479201331114809, + "grad_norm": 1.1702793360559631, + "learning_rate": 9.767038621379681e-06, + "loss": 0.6491, + "step": 525 + }, + { + "epoch": 0.1250297123841217, + "grad_norm": 1.3505975655768538, + "learning_rate": 9.765875724594913e-06, + "loss": 0.6655, + "step": 526 + }, + { + "epoch": 0.12526741145709533, + "grad_norm": 1.3790353721912545, + "learning_rate": 9.764710002135784e-06, + "loss": 0.6461, + "step": 527 + }, + { + "epoch": 0.12550511053006894, + "grad_norm": 1.339746215805567, + "learning_rate": 9.763541454693445e-06, + "loss": 0.7024, + "step": 528 + }, + { + "epoch": 0.12574280960304254, + "grad_norm": 1.221539734372046, + "learning_rate": 9.762370082960727e-06, + "loss": 0.7004, + "step": 529 + }, + { + "epoch": 0.12598050867601615, + "grad_norm": 1.2375325216050186, + "learning_rate": 9.761195887632131e-06, + "loss": 0.703, + "step": 530 + }, + { + "epoch": 0.12621820774898979, + "grad_norm": 1.3345059101370589, + "learning_rate": 9.760018869403837e-06, + "loss": 0.7209, + "step": 531 + }, + { + "epoch": 0.1264559068219634, + "grad_norm": 1.4054967822024778, + "learning_rate": 9.758839028973693e-06, + "loss": 0.6657, + "step": 532 + }, + { + "epoch": 0.126693605894937, + "grad_norm": 1.5548205023747792, + "learning_rate": 9.757656367041222e-06, + "loss": 0.6786, + "step": 533 + }, + { + "epoch": 0.12693130496791064, + "grad_norm": 1.3135651930787553, + "learning_rate": 9.756470884307625e-06, + "loss": 0.6943, + "step": 534 + }, + { + "epoch": 0.12716900404088424, + "grad_norm": 1.420851259897442, + "learning_rate": 9.755282581475769e-06, + "loss": 0.6765, + "step": 535 + }, + { + "epoch": 0.12740670311385785, + "grad_norm": 1.4790310581032111, + "learning_rate": 9.754091459250194e-06, + "loss": 0.6236, + "step": 536 + }, + { + "epoch": 0.12764440218683148, + "grad_norm": 1.303734587382992, + "learning_rate": 9.752897518337117e-06, + "loss": 0.6768, + "step": 537 + }, + { + "epoch": 0.1278821012598051, + "grad_norm": 1.169118652412735, + "learning_rate": 9.75170075944442e-06, + "loss": 0.6888, + "step": 538 + }, + { + "epoch": 0.1281198003327787, + "grad_norm": 1.6842536125214178, + "learning_rate": 9.750501183281655e-06, + "loss": 0.6762, + "step": 539 + }, + { + "epoch": 0.1283574994057523, + "grad_norm": 1.0567822345937925, + "learning_rate": 9.749298790560054e-06, + "loss": 0.6773, + "step": 540 + }, + { + "epoch": 0.12859519847872594, + "grad_norm": 1.177357832306313, + "learning_rate": 9.748093581992506e-06, + "loss": 0.6806, + "step": 541 + }, + { + "epoch": 0.12883289755169955, + "grad_norm": 1.9677643603718509, + "learning_rate": 9.74688555829358e-06, + "loss": 0.706, + "step": 542 + }, + { + "epoch": 0.12907059662467316, + "grad_norm": 1.1759388924558996, + "learning_rate": 9.745674720179507e-06, + "loss": 0.7111, + "step": 543 + }, + { + "epoch": 0.1293082956976468, + "grad_norm": 1.278756606661313, + "learning_rate": 9.744461068368191e-06, + "loss": 0.6884, + "step": 544 + }, + { + "epoch": 0.1295459947706204, + "grad_norm": 1.1917426927284331, + "learning_rate": 9.743244603579202e-06, + "loss": 0.6776, + "step": 545 + }, + { + "epoch": 0.129783693843594, + "grad_norm": 1.2549288175373294, + "learning_rate": 9.74202532653378e-06, + "loss": 0.6415, + "step": 546 + }, + { + "epoch": 0.1300213929165676, + "grad_norm": 1.312271691653255, + "learning_rate": 9.74080323795483e-06, + "loss": 0.6087, + "step": 547 + }, + { + "epoch": 0.13025909198954125, + "grad_norm": 1.2834520355675827, + "learning_rate": 9.739578338566923e-06, + "loss": 0.6547, + "step": 548 + }, + { + "epoch": 0.13049679106251486, + "grad_norm": 1.307883376386262, + "learning_rate": 9.738350629096304e-06, + "loss": 0.7142, + "step": 549 + }, + { + "epoch": 0.13073449013548846, + "grad_norm": 1.1707554705693886, + "learning_rate": 9.737120110270872e-06, + "loss": 0.7354, + "step": 550 + }, + { + "epoch": 0.1309721892084621, + "grad_norm": 1.2640272805357968, + "learning_rate": 9.735886782820202e-06, + "loss": 0.6547, + "step": 551 + }, + { + "epoch": 0.1312098882814357, + "grad_norm": 1.1171242765035656, + "learning_rate": 9.73465064747553e-06, + "loss": 0.6547, + "step": 552 + }, + { + "epoch": 0.1314475873544093, + "grad_norm": 1.3925144114092014, + "learning_rate": 9.733411704969754e-06, + "loss": 0.64, + "step": 553 + }, + { + "epoch": 0.13168528642738292, + "grad_norm": 1.2229378187031883, + "learning_rate": 9.732169956037444e-06, + "loss": 0.6912, + "step": 554 + }, + { + "epoch": 0.13192298550035655, + "grad_norm": 1.3153958508100339, + "learning_rate": 9.730925401414828e-06, + "loss": 0.6854, + "step": 555 + }, + { + "epoch": 0.13216068457333016, + "grad_norm": 1.3158585096546145, + "learning_rate": 9.729678041839798e-06, + "loss": 0.6325, + "step": 556 + }, + { + "epoch": 0.13239838364630377, + "grad_norm": 1.2770654323885802, + "learning_rate": 9.72842787805191e-06, + "loss": 0.6924, + "step": 557 + }, + { + "epoch": 0.1326360827192774, + "grad_norm": 1.6199283845321133, + "learning_rate": 9.727174910792383e-06, + "loss": 0.6619, + "step": 558 + }, + { + "epoch": 0.132873781792251, + "grad_norm": 1.2195753386769346, + "learning_rate": 9.7259191408041e-06, + "loss": 0.6325, + "step": 559 + }, + { + "epoch": 0.13311148086522462, + "grad_norm": 1.1626201540966494, + "learning_rate": 9.724660568831599e-06, + "loss": 0.6411, + "step": 560 + }, + { + "epoch": 0.13334917993819825, + "grad_norm": 1.4220424352489112, + "learning_rate": 9.723399195621087e-06, + "loss": 0.7205, + "step": 561 + }, + { + "epoch": 0.13358687901117186, + "grad_norm": 1.293136944674464, + "learning_rate": 9.722135021920427e-06, + "loss": 0.6958, + "step": 562 + }, + { + "epoch": 0.13382457808414547, + "grad_norm": 2.5186632031721796, + "learning_rate": 9.720868048479146e-06, + "loss": 0.7479, + "step": 563 + }, + { + "epoch": 0.13406227715711908, + "grad_norm": 1.1990225156502945, + "learning_rate": 9.719598276048426e-06, + "loss": 0.6396, + "step": 564 + }, + { + "epoch": 0.1342999762300927, + "grad_norm": 1.1934513491452168, + "learning_rate": 9.718325705381115e-06, + "loss": 0.6675, + "step": 565 + }, + { + "epoch": 0.13453767530306632, + "grad_norm": 1.1957132532118826, + "learning_rate": 9.717050337231715e-06, + "loss": 0.6058, + "step": 566 + }, + { + "epoch": 0.13477537437603992, + "grad_norm": 1.1785407504523764, + "learning_rate": 9.715772172356388e-06, + "loss": 0.6622, + "step": 567 + }, + { + "epoch": 0.13501307344901356, + "grad_norm": 1.3057851903965536, + "learning_rate": 9.714491211512955e-06, + "loss": 0.7086, + "step": 568 + }, + { + "epoch": 0.13525077252198717, + "grad_norm": 1.4488125707922088, + "learning_rate": 9.713207455460893e-06, + "loss": 0.7, + "step": 569 + }, + { + "epoch": 0.13548847159496077, + "grad_norm": 1.194283723589389, + "learning_rate": 9.711920904961338e-06, + "loss": 0.6826, + "step": 570 + }, + { + "epoch": 0.13572617066793438, + "grad_norm": 1.1968886305408326, + "learning_rate": 9.710631560777082e-06, + "loss": 0.6119, + "step": 571 + }, + { + "epoch": 0.13596386974090802, + "grad_norm": 1.2458620564850562, + "learning_rate": 9.709339423672575e-06, + "loss": 0.6901, + "step": 572 + }, + { + "epoch": 0.13620156881388162, + "grad_norm": 1.208425932789428, + "learning_rate": 9.708044494413919e-06, + "loss": 0.7159, + "step": 573 + }, + { + "epoch": 0.13643926788685523, + "grad_norm": 1.2410672227057697, + "learning_rate": 9.706746773768875e-06, + "loss": 0.6634, + "step": 574 + }, + { + "epoch": 0.13667696695982887, + "grad_norm": 1.3836965994521226, + "learning_rate": 9.705446262506858e-06, + "loss": 0.6566, + "step": 575 + }, + { + "epoch": 0.13691466603280247, + "grad_norm": 1.7019874771595633, + "learning_rate": 9.704142961398938e-06, + "loss": 0.6229, + "step": 576 + }, + { + "epoch": 0.13715236510577608, + "grad_norm": 1.5633864663861765, + "learning_rate": 9.702836871217838e-06, + "loss": 0.662, + "step": 577 + }, + { + "epoch": 0.13739006417874972, + "grad_norm": 1.2296307213834012, + "learning_rate": 9.701527992737936e-06, + "loss": 0.7195, + "step": 578 + }, + { + "epoch": 0.13762776325172332, + "grad_norm": 1.2107372526160358, + "learning_rate": 9.700216326735263e-06, + "loss": 0.6249, + "step": 579 + }, + { + "epoch": 0.13786546232469693, + "grad_norm": 1.2628248537708961, + "learning_rate": 9.698901873987498e-06, + "loss": 0.6208, + "step": 580 + }, + { + "epoch": 0.13810316139767054, + "grad_norm": 1.775843463841929, + "learning_rate": 9.69758463527398e-06, + "loss": 0.6928, + "step": 581 + }, + { + "epoch": 0.13834086047064417, + "grad_norm": 1.2296282359700814, + "learning_rate": 9.696264611375697e-06, + "loss": 0.6988, + "step": 582 + }, + { + "epoch": 0.13857855954361778, + "grad_norm": 1.2790772106879231, + "learning_rate": 9.694941803075285e-06, + "loss": 0.6522, + "step": 583 + }, + { + "epoch": 0.1388162586165914, + "grad_norm": 1.2131433002299092, + "learning_rate": 9.693616211157032e-06, + "loss": 0.6794, + "step": 584 + }, + { + "epoch": 0.13905395768956502, + "grad_norm": 1.1869478600639272, + "learning_rate": 9.692287836406881e-06, + "loss": 0.6733, + "step": 585 + }, + { + "epoch": 0.13929165676253863, + "grad_norm": 1.381713031758582, + "learning_rate": 9.690956679612422e-06, + "loss": 0.6565, + "step": 586 + }, + { + "epoch": 0.13952935583551224, + "grad_norm": 1.2630731101811503, + "learning_rate": 9.689622741562891e-06, + "loss": 0.6586, + "step": 587 + }, + { + "epoch": 0.13976705490848584, + "grad_norm": 1.1758158998608659, + "learning_rate": 9.688286023049178e-06, + "loss": 0.678, + "step": 588 + }, + { + "epoch": 0.14000475398145948, + "grad_norm": 1.3420505039457131, + "learning_rate": 9.686946524863821e-06, + "loss": 0.695, + "step": 589 + }, + { + "epoch": 0.14024245305443309, + "grad_norm": 1.07386422202855, + "learning_rate": 9.685604247801002e-06, + "loss": 0.6509, + "step": 590 + }, + { + "epoch": 0.1404801521274067, + "grad_norm": 1.2863203294949979, + "learning_rate": 9.684259192656554e-06, + "loss": 0.6533, + "step": 591 + }, + { + "epoch": 0.14071785120038033, + "grad_norm": 1.3808296006379102, + "learning_rate": 9.682911360227958e-06, + "loss": 0.713, + "step": 592 + }, + { + "epoch": 0.14095555027335394, + "grad_norm": 1.3921657671149494, + "learning_rate": 9.681560751314338e-06, + "loss": 0.6591, + "step": 593 + }, + { + "epoch": 0.14119324934632754, + "grad_norm": 1.193754628852629, + "learning_rate": 9.680207366716467e-06, + "loss": 0.6743, + "step": 594 + }, + { + "epoch": 0.14143094841930118, + "grad_norm": 1.3218346236441667, + "learning_rate": 9.678851207236764e-06, + "loss": 0.599, + "step": 595 + }, + { + "epoch": 0.14166864749227479, + "grad_norm": 1.2920960495468623, + "learning_rate": 9.677492273679291e-06, + "loss": 0.7113, + "step": 596 + }, + { + "epoch": 0.1419063465652484, + "grad_norm": 1.7995562615958423, + "learning_rate": 9.676130566849757e-06, + "loss": 0.6988, + "step": 597 + }, + { + "epoch": 0.142144045638222, + "grad_norm": 1.4718021091974418, + "learning_rate": 9.674766087555512e-06, + "loss": 0.6807, + "step": 598 + }, + { + "epoch": 0.14238174471119563, + "grad_norm": 1.263698973560867, + "learning_rate": 9.673398836605554e-06, + "loss": 0.6401, + "step": 599 + }, + { + "epoch": 0.14261944378416924, + "grad_norm": 1.1844020937948099, + "learning_rate": 9.672028814810521e-06, + "loss": 0.698, + "step": 600 + }, + { + "epoch": 0.14261944378416924, + "eval_loss": 0.6724269986152649, + "eval_runtime": 277.9303, + "eval_samples_per_second": 39.139, + "eval_steps_per_second": 1.223, + "step": 600 + }, + { + "epoch": 0.14285714285714285, + "grad_norm": 1.2148283085993112, + "learning_rate": 9.670656022982696e-06, + "loss": 0.6205, + "step": 601 + }, + { + "epoch": 0.14309484193011648, + "grad_norm": 1.3222187633955464, + "learning_rate": 9.669280461936004e-06, + "loss": 0.6861, + "step": 602 + }, + { + "epoch": 0.1433325410030901, + "grad_norm": 1.5063295373753822, + "learning_rate": 9.667902132486009e-06, + "loss": 0.7282, + "step": 603 + }, + { + "epoch": 0.1435702400760637, + "grad_norm": 1.417719503354877, + "learning_rate": 9.66652103544992e-06, + "loss": 0.6716, + "step": 604 + }, + { + "epoch": 0.1438079391490373, + "grad_norm": 1.2059327003670695, + "learning_rate": 9.665137171646588e-06, + "loss": 0.6382, + "step": 605 + }, + { + "epoch": 0.14404563822201094, + "grad_norm": 1.251380560441615, + "learning_rate": 9.663750541896498e-06, + "loss": 0.6614, + "step": 606 + }, + { + "epoch": 0.14428333729498455, + "grad_norm": 1.285241943734605, + "learning_rate": 9.66236114702178e-06, + "loss": 0.6745, + "step": 607 + }, + { + "epoch": 0.14452103636795816, + "grad_norm": 1.24185601686558, + "learning_rate": 9.660968987846203e-06, + "loss": 0.7248, + "step": 608 + }, + { + "epoch": 0.1447587354409318, + "grad_norm": 1.940905662630259, + "learning_rate": 9.659574065195174e-06, + "loss": 0.6757, + "step": 609 + }, + { + "epoch": 0.1449964345139054, + "grad_norm": 1.7266030862623274, + "learning_rate": 9.65817637989574e-06, + "loss": 0.6976, + "step": 610 + }, + { + "epoch": 0.145234133586879, + "grad_norm": 1.180457793499378, + "learning_rate": 9.656775932776582e-06, + "loss": 0.6578, + "step": 611 + }, + { + "epoch": 0.14547183265985264, + "grad_norm": 1.3720375844236272, + "learning_rate": 9.655372724668022e-06, + "loss": 0.6904, + "step": 612 + }, + { + "epoch": 0.14570953173282625, + "grad_norm": 1.2970209927480707, + "learning_rate": 9.65396675640202e-06, + "loss": 0.6575, + "step": 613 + }, + { + "epoch": 0.14594723080579985, + "grad_norm": 2.1475566913602093, + "learning_rate": 9.652558028812171e-06, + "loss": 0.7157, + "step": 614 + }, + { + "epoch": 0.14618492987877346, + "grad_norm": 1.2224400223053387, + "learning_rate": 9.651146542733702e-06, + "loss": 0.6739, + "step": 615 + }, + { + "epoch": 0.1464226289517471, + "grad_norm": 1.3201692400597504, + "learning_rate": 9.649732299003482e-06, + "loss": 0.6699, + "step": 616 + }, + { + "epoch": 0.1466603280247207, + "grad_norm": 2.4988392670723507, + "learning_rate": 9.64831529846001e-06, + "loss": 0.6893, + "step": 617 + }, + { + "epoch": 0.1468980270976943, + "grad_norm": 1.305479125625365, + "learning_rate": 9.646895541943427e-06, + "loss": 0.6824, + "step": 618 + }, + { + "epoch": 0.14713572617066795, + "grad_norm": 1.2199698856792385, + "learning_rate": 9.645473030295496e-06, + "loss": 0.6876, + "step": 619 + }, + { + "epoch": 0.14737342524364155, + "grad_norm": 1.235595815456883, + "learning_rate": 9.644047764359623e-06, + "loss": 0.7255, + "step": 620 + }, + { + "epoch": 0.14761112431661516, + "grad_norm": 1.2470114163965813, + "learning_rate": 9.642619744980844e-06, + "loss": 0.6668, + "step": 621 + }, + { + "epoch": 0.14784882338958877, + "grad_norm": 1.2553389046321473, + "learning_rate": 9.641188973005826e-06, + "loss": 0.6661, + "step": 622 + }, + { + "epoch": 0.1480865224625624, + "grad_norm": 1.232123019845644, + "learning_rate": 9.639755449282874e-06, + "loss": 0.6864, + "step": 623 + }, + { + "epoch": 0.148324221535536, + "grad_norm": 1.5995635505799761, + "learning_rate": 9.638319174661916e-06, + "loss": 0.6716, + "step": 624 + }, + { + "epoch": 0.14856192060850962, + "grad_norm": 1.1539280535939245, + "learning_rate": 9.636880149994518e-06, + "loss": 0.6105, + "step": 625 + }, + { + "epoch": 0.14879961968148325, + "grad_norm": 1.1870525383115742, + "learning_rate": 9.635438376133872e-06, + "loss": 0.7117, + "step": 626 + }, + { + "epoch": 0.14903731875445686, + "grad_norm": 1.26773807236903, + "learning_rate": 9.633993853934803e-06, + "loss": 0.5916, + "step": 627 + }, + { + "epoch": 0.14927501782743047, + "grad_norm": 1.296037934124825, + "learning_rate": 9.632546584253763e-06, + "loss": 0.6513, + "step": 628 + }, + { + "epoch": 0.14951271690040407, + "grad_norm": 1.2966940063968515, + "learning_rate": 9.631096567948834e-06, + "loss": 0.6532, + "step": 629 + }, + { + "epoch": 0.1497504159733777, + "grad_norm": 1.2990110920495674, + "learning_rate": 9.629643805879729e-06, + "loss": 0.7202, + "step": 630 + }, + { + "epoch": 0.14998811504635132, + "grad_norm": 1.4470463289035786, + "learning_rate": 9.628188298907782e-06, + "loss": 0.6646, + "step": 631 + }, + { + "epoch": 0.15022581411932492, + "grad_norm": 1.59062299023798, + "learning_rate": 9.626730047895962e-06, + "loss": 0.6497, + "step": 632 + }, + { + "epoch": 0.15046351319229856, + "grad_norm": 1.212794040990129, + "learning_rate": 9.625269053708863e-06, + "loss": 0.6483, + "step": 633 + }, + { + "epoch": 0.15070121226527217, + "grad_norm": 1.2423796173720432, + "learning_rate": 9.623805317212703e-06, + "loss": 0.6408, + "step": 634 + }, + { + "epoch": 0.15093891133824577, + "grad_norm": 1.408440793372717, + "learning_rate": 9.622338839275327e-06, + "loss": 0.6493, + "step": 635 + }, + { + "epoch": 0.1511766104112194, + "grad_norm": 1.1905408661741073, + "learning_rate": 9.620869620766205e-06, + "loss": 0.6585, + "step": 636 + }, + { + "epoch": 0.15141430948419302, + "grad_norm": 1.2563057578685959, + "learning_rate": 9.619397662556434e-06, + "loss": 0.6766, + "step": 637 + }, + { + "epoch": 0.15165200855716662, + "grad_norm": 1.2829837226360898, + "learning_rate": 9.617922965518733e-06, + "loss": 0.6954, + "step": 638 + }, + { + "epoch": 0.15188970763014023, + "grad_norm": 1.4038777675245564, + "learning_rate": 9.616445530527448e-06, + "loss": 0.6441, + "step": 639 + }, + { + "epoch": 0.15212740670311387, + "grad_norm": 1.100133058722965, + "learning_rate": 9.614965358458542e-06, + "loss": 0.6481, + "step": 640 + }, + { + "epoch": 0.15236510577608747, + "grad_norm": 1.4064134795822976, + "learning_rate": 9.613482450189608e-06, + "loss": 0.7099, + "step": 641 + }, + { + "epoch": 0.15260280484906108, + "grad_norm": 1.1958601631219874, + "learning_rate": 9.611996806599858e-06, + "loss": 0.7181, + "step": 642 + }, + { + "epoch": 0.15284050392203472, + "grad_norm": 1.2010436254204155, + "learning_rate": 9.610508428570122e-06, + "loss": 0.691, + "step": 643 + }, + { + "epoch": 0.15307820299500832, + "grad_norm": 1.3669689079825076, + "learning_rate": 9.60901731698286e-06, + "loss": 0.6612, + "step": 644 + }, + { + "epoch": 0.15331590206798193, + "grad_norm": 1.1868309339050405, + "learning_rate": 9.607523472722148e-06, + "loss": 0.6571, + "step": 645 + }, + { + "epoch": 0.15355360114095554, + "grad_norm": 1.2463665774830837, + "learning_rate": 9.60602689667368e-06, + "loss": 0.686, + "step": 646 + }, + { + "epoch": 0.15379130021392917, + "grad_norm": 3.043234131082634, + "learning_rate": 9.60452758972477e-06, + "loss": 0.7425, + "step": 647 + }, + { + "epoch": 0.15402899928690278, + "grad_norm": 1.3305797430740234, + "learning_rate": 9.603025552764355e-06, + "loss": 0.6773, + "step": 648 + }, + { + "epoch": 0.1542666983598764, + "grad_norm": 1.4198260876690318, + "learning_rate": 9.601520786682989e-06, + "loss": 0.66, + "step": 649 + }, + { + "epoch": 0.15450439743285002, + "grad_norm": 1.5323211118569264, + "learning_rate": 9.600013292372843e-06, + "loss": 0.6957, + "step": 650 + }, + { + "epoch": 0.15474209650582363, + "grad_norm": 1.3152996387248488, + "learning_rate": 9.598503070727705e-06, + "loss": 0.6268, + "step": 651 + }, + { + "epoch": 0.15497979557879724, + "grad_norm": 1.2403114725246491, + "learning_rate": 9.596990122642984e-06, + "loss": 0.6923, + "step": 652 + }, + { + "epoch": 0.15521749465177087, + "grad_norm": 1.5220891461879849, + "learning_rate": 9.5954744490157e-06, + "loss": 0.6567, + "step": 653 + }, + { + "epoch": 0.15545519372474448, + "grad_norm": 1.436209648281884, + "learning_rate": 9.593956050744493e-06, + "loss": 0.6503, + "step": 654 + }, + { + "epoch": 0.15569289279771809, + "grad_norm": 1.4749607887564242, + "learning_rate": 9.592434928729617e-06, + "loss": 0.6601, + "step": 655 + }, + { + "epoch": 0.1559305918706917, + "grad_norm": 1.3996681655335288, + "learning_rate": 9.59091108387294e-06, + "loss": 0.6451, + "step": 656 + }, + { + "epoch": 0.15616829094366533, + "grad_norm": 1.327095108508895, + "learning_rate": 9.589384517077945e-06, + "loss": 0.6742, + "step": 657 + }, + { + "epoch": 0.15640599001663893, + "grad_norm": 1.0909232372611186, + "learning_rate": 9.587855229249732e-06, + "loss": 0.6705, + "step": 658 + }, + { + "epoch": 0.15664368908961254, + "grad_norm": 1.2944459972301818, + "learning_rate": 9.58632322129501e-06, + "loss": 0.6888, + "step": 659 + }, + { + "epoch": 0.15688138816258618, + "grad_norm": 1.4249873323999789, + "learning_rate": 9.5847884941221e-06, + "loss": 0.6844, + "step": 660 + }, + { + "epoch": 0.15711908723555978, + "grad_norm": 1.2029715595747381, + "learning_rate": 9.583251048640941e-06, + "loss": 0.7095, + "step": 661 + }, + { + "epoch": 0.1573567863085334, + "grad_norm": 1.4255948458859287, + "learning_rate": 9.581710885763077e-06, + "loss": 0.6472, + "step": 662 + }, + { + "epoch": 0.157594485381507, + "grad_norm": 1.4052535031860172, + "learning_rate": 9.58016800640167e-06, + "loss": 0.6486, + "step": 663 + }, + { + "epoch": 0.15783218445448063, + "grad_norm": 2.3363886079600436, + "learning_rate": 9.578622411471482e-06, + "loss": 0.7074, + "step": 664 + }, + { + "epoch": 0.15806988352745424, + "grad_norm": 1.2653874658529207, + "learning_rate": 9.577074101888901e-06, + "loss": 0.6413, + "step": 665 + }, + { + "epoch": 0.15830758260042785, + "grad_norm": 1.4045419961749064, + "learning_rate": 9.575523078571909e-06, + "loss": 0.6212, + "step": 666 + }, + { + "epoch": 0.15854528167340148, + "grad_norm": 1.2126819374647295, + "learning_rate": 9.573969342440107e-06, + "loss": 0.6888, + "step": 667 + }, + { + "epoch": 0.1587829807463751, + "grad_norm": 1.3545552778270544, + "learning_rate": 9.572412894414696e-06, + "loss": 0.6355, + "step": 668 + }, + { + "epoch": 0.1590206798193487, + "grad_norm": 1.5426723910113167, + "learning_rate": 9.570853735418494e-06, + "loss": 0.6654, + "step": 669 + }, + { + "epoch": 0.15925837889232233, + "grad_norm": 1.279513157084368, + "learning_rate": 9.56929186637592e-06, + "loss": 0.7181, + "step": 670 + }, + { + "epoch": 0.15949607796529594, + "grad_norm": 1.4592233534214594, + "learning_rate": 9.567727288213005e-06, + "loss": 0.6843, + "step": 671 + }, + { + "epoch": 0.15973377703826955, + "grad_norm": 1.3453703325797093, + "learning_rate": 9.566160001857378e-06, + "loss": 0.6512, + "step": 672 + }, + { + "epoch": 0.15997147611124315, + "grad_norm": 1.348687715871197, + "learning_rate": 9.564590008238284e-06, + "loss": 0.7001, + "step": 673 + }, + { + "epoch": 0.1602091751842168, + "grad_norm": 1.2066978961690804, + "learning_rate": 9.563017308286563e-06, + "loss": 0.64, + "step": 674 + }, + { + "epoch": 0.1604468742571904, + "grad_norm": 1.4595058570271036, + "learning_rate": 9.561441902934667e-06, + "loss": 0.6723, + "step": 675 + }, + { + "epoch": 0.160684573330164, + "grad_norm": 1.1226806914522804, + "learning_rate": 9.559863793116647e-06, + "loss": 0.6425, + "step": 676 + }, + { + "epoch": 0.16092227240313764, + "grad_norm": 1.2418456878057371, + "learning_rate": 9.558282979768164e-06, + "loss": 0.6451, + "step": 677 + }, + { + "epoch": 0.16115997147611125, + "grad_norm": 1.281978190342555, + "learning_rate": 9.556699463826475e-06, + "loss": 0.6214, + "step": 678 + }, + { + "epoch": 0.16139767054908485, + "grad_norm": 1.5690617104245173, + "learning_rate": 9.555113246230443e-06, + "loss": 0.6792, + "step": 679 + }, + { + "epoch": 0.16163536962205846, + "grad_norm": 1.0982176613442869, + "learning_rate": 9.553524327920532e-06, + "loss": 0.6278, + "step": 680 + }, + { + "epoch": 0.1618730686950321, + "grad_norm": 1.3112873961066087, + "learning_rate": 9.551932709838806e-06, + "loss": 0.6765, + "step": 681 + }, + { + "epoch": 0.1621107677680057, + "grad_norm": 1.6415396837273415, + "learning_rate": 9.550338392928931e-06, + "loss": 0.6718, + "step": 682 + }, + { + "epoch": 0.1623484668409793, + "grad_norm": 1.3699930772178417, + "learning_rate": 9.548741378136173e-06, + "loss": 0.6927, + "step": 683 + }, + { + "epoch": 0.16258616591395295, + "grad_norm": 1.2036279048161975, + "learning_rate": 9.5471416664074e-06, + "loss": 0.6499, + "step": 684 + }, + { + "epoch": 0.16282386498692655, + "grad_norm": 1.1490543591250955, + "learning_rate": 9.545539258691076e-06, + "loss": 0.6583, + "step": 685 + }, + { + "epoch": 0.16306156405990016, + "grad_norm": 1.2044408707404983, + "learning_rate": 9.543934155937261e-06, + "loss": 0.6266, + "step": 686 + }, + { + "epoch": 0.1632992631328738, + "grad_norm": 1.3422530202717076, + "learning_rate": 9.542326359097619e-06, + "loss": 0.68, + "step": 687 + }, + { + "epoch": 0.1635369622058474, + "grad_norm": 1.3841861682581356, + "learning_rate": 9.540715869125407e-06, + "loss": 0.6508, + "step": 688 + }, + { + "epoch": 0.163774661278821, + "grad_norm": 1.278602041339616, + "learning_rate": 9.539102686975481e-06, + "loss": 0.618, + "step": 689 + }, + { + "epoch": 0.16401236035179462, + "grad_norm": 1.4094830194583823, + "learning_rate": 9.53748681360429e-06, + "loss": 0.6601, + "step": 690 + }, + { + "epoch": 0.16425005942476825, + "grad_norm": 1.3202254631450843, + "learning_rate": 9.535868249969882e-06, + "loss": 0.6529, + "step": 691 + }, + { + "epoch": 0.16448775849774186, + "grad_norm": 1.4239668483803989, + "learning_rate": 9.534246997031901e-06, + "loss": 0.6915, + "step": 692 + }, + { + "epoch": 0.16472545757071547, + "grad_norm": 1.6385481770055847, + "learning_rate": 9.53262305575158e-06, + "loss": 0.6475, + "step": 693 + }, + { + "epoch": 0.1649631566436891, + "grad_norm": 1.197549613704571, + "learning_rate": 9.53099642709175e-06, + "loss": 0.6467, + "step": 694 + }, + { + "epoch": 0.1652008557166627, + "grad_norm": 1.132614896880335, + "learning_rate": 9.529367112016836e-06, + "loss": 0.6744, + "step": 695 + }, + { + "epoch": 0.16543855478963632, + "grad_norm": 1.1511542550970815, + "learning_rate": 9.527735111492852e-06, + "loss": 0.6633, + "step": 696 + }, + { + "epoch": 0.16567625386260992, + "grad_norm": 1.2857711466240023, + "learning_rate": 9.52610042648741e-06, + "loss": 0.7178, + "step": 697 + }, + { + "epoch": 0.16591395293558356, + "grad_norm": 1.3244338476587183, + "learning_rate": 9.524463057969709e-06, + "loss": 0.65, + "step": 698 + }, + { + "epoch": 0.16615165200855717, + "grad_norm": 1.3276329883818778, + "learning_rate": 9.522823006910538e-06, + "loss": 0.6506, + "step": 699 + }, + { + "epoch": 0.16638935108153077, + "grad_norm": 1.7331728036967384, + "learning_rate": 9.521180274282279e-06, + "loss": 0.6794, + "step": 700 + }, + { + "epoch": 0.16638935108153077, + "eval_loss": 0.6647465825080872, + "eval_runtime": 284.7475, + "eval_samples_per_second": 38.202, + "eval_steps_per_second": 1.194, + "step": 700 + }, + { + "epoch": 0.1666270501545044, + "grad_norm": 1.1764018589232856, + "learning_rate": 9.519534861058905e-06, + "loss": 0.6697, + "step": 701 + }, + { + "epoch": 0.16686474922747802, + "grad_norm": 1.1024973031028107, + "learning_rate": 9.517886768215978e-06, + "loss": 0.6236, + "step": 702 + }, + { + "epoch": 0.16710244830045162, + "grad_norm": 1.242853483258734, + "learning_rate": 9.516235996730645e-06, + "loss": 0.6524, + "step": 703 + }, + { + "epoch": 0.16734014737342523, + "grad_norm": 1.1750142505825112, + "learning_rate": 9.514582547581646e-06, + "loss": 0.5965, + "step": 704 + }, + { + "epoch": 0.16757784644639886, + "grad_norm": 1.6852829010582093, + "learning_rate": 9.512926421749305e-06, + "loss": 0.6211, + "step": 705 + }, + { + "epoch": 0.16781554551937247, + "grad_norm": 1.5141872448389566, + "learning_rate": 9.511267620215533e-06, + "loss": 0.6412, + "step": 706 + }, + { + "epoch": 0.16805324459234608, + "grad_norm": 1.4663518519510934, + "learning_rate": 9.509606143963832e-06, + "loss": 0.6472, + "step": 707 + }, + { + "epoch": 0.16829094366531971, + "grad_norm": 1.5892550590249068, + "learning_rate": 9.507941993979284e-06, + "loss": 0.7493, + "step": 708 + }, + { + "epoch": 0.16852864273829332, + "grad_norm": 1.3739591729970622, + "learning_rate": 9.50627517124856e-06, + "loss": 0.6333, + "step": 709 + }, + { + "epoch": 0.16876634181126693, + "grad_norm": 1.5223061304755066, + "learning_rate": 9.504605676759917e-06, + "loss": 0.6492, + "step": 710 + }, + { + "epoch": 0.16900404088424056, + "grad_norm": 1.4348929963195451, + "learning_rate": 9.502933511503187e-06, + "loss": 0.7214, + "step": 711 + }, + { + "epoch": 0.16924173995721417, + "grad_norm": 1.4590315065967934, + "learning_rate": 9.5012586764698e-06, + "loss": 0.6614, + "step": 712 + }, + { + "epoch": 0.16947943903018778, + "grad_norm": 1.4478104704925558, + "learning_rate": 9.499581172652754e-06, + "loss": 0.5815, + "step": 713 + }, + { + "epoch": 0.16971713810316139, + "grad_norm": 1.2524132982278728, + "learning_rate": 9.497901001046642e-06, + "loss": 0.6594, + "step": 714 + }, + { + "epoch": 0.16995483717613502, + "grad_norm": 1.22042514106662, + "learning_rate": 9.496218162647629e-06, + "loss": 0.6204, + "step": 715 + }, + { + "epoch": 0.17019253624910863, + "grad_norm": 1.7092692113634425, + "learning_rate": 9.494532658453467e-06, + "loss": 0.7013, + "step": 716 + }, + { + "epoch": 0.17043023532208224, + "grad_norm": 1.6428069764731597, + "learning_rate": 9.492844489463486e-06, + "loss": 0.6621, + "step": 717 + }, + { + "epoch": 0.17066793439505587, + "grad_norm": 1.3763734730090518, + "learning_rate": 9.491153656678597e-06, + "loss": 0.6264, + "step": 718 + }, + { + "epoch": 0.17090563346802948, + "grad_norm": 1.3580994331782468, + "learning_rate": 9.489460161101291e-06, + "loss": 0.6865, + "step": 719 + }, + { + "epoch": 0.17114333254100308, + "grad_norm": 1.2227734741671612, + "learning_rate": 9.487764003735636e-06, + "loss": 0.67, + "step": 720 + }, + { + "epoch": 0.1713810316139767, + "grad_norm": 1.1608736519845244, + "learning_rate": 9.486065185587278e-06, + "loss": 0.5981, + "step": 721 + }, + { + "epoch": 0.17161873068695033, + "grad_norm": 1.249031351233754, + "learning_rate": 9.484363707663443e-06, + "loss": 0.6529, + "step": 722 + }, + { + "epoch": 0.17185642975992393, + "grad_norm": 1.3975394570210213, + "learning_rate": 9.48265957097293e-06, + "loss": 0.6508, + "step": 723 + }, + { + "epoch": 0.17209412883289754, + "grad_norm": 1.1431005873008482, + "learning_rate": 9.48095277652612e-06, + "loss": 0.6329, + "step": 724 + }, + { + "epoch": 0.17233182790587118, + "grad_norm": 1.220767571637931, + "learning_rate": 9.479243325334965e-06, + "loss": 0.7053, + "step": 725 + }, + { + "epoch": 0.17256952697884478, + "grad_norm": 1.3697293553945997, + "learning_rate": 9.477531218412995e-06, + "loss": 0.6349, + "step": 726 + }, + { + "epoch": 0.1728072260518184, + "grad_norm": 1.401760925324015, + "learning_rate": 9.475816456775313e-06, + "loss": 0.6617, + "step": 727 + }, + { + "epoch": 0.17304492512479203, + "grad_norm": 1.337875169047814, + "learning_rate": 9.474099041438594e-06, + "loss": 0.6989, + "step": 728 + }, + { + "epoch": 0.17328262419776563, + "grad_norm": 1.1904650460892745, + "learning_rate": 9.472378973421092e-06, + "loss": 0.6321, + "step": 729 + }, + { + "epoch": 0.17352032327073924, + "grad_norm": 1.3145615408934288, + "learning_rate": 9.47065625374263e-06, + "loss": 0.6753, + "step": 730 + }, + { + "epoch": 0.17375802234371285, + "grad_norm": 2.2583545506768927, + "learning_rate": 9.4689308834246e-06, + "loss": 0.6282, + "step": 731 + }, + { + "epoch": 0.17399572141668648, + "grad_norm": 1.49974607738386, + "learning_rate": 9.467202863489974e-06, + "loss": 0.6056, + "step": 732 + }, + { + "epoch": 0.1742334204896601, + "grad_norm": 1.3618290061685783, + "learning_rate": 9.465472194963287e-06, + "loss": 0.6271, + "step": 733 + }, + { + "epoch": 0.1744711195626337, + "grad_norm": 1.4499198493159817, + "learning_rate": 9.46373887887065e-06, + "loss": 0.701, + "step": 734 + }, + { + "epoch": 0.17470881863560733, + "grad_norm": 1.3929436492227258, + "learning_rate": 9.46200291623974e-06, + "loss": 0.6519, + "step": 735 + }, + { + "epoch": 0.17494651770858094, + "grad_norm": 1.7706890321912825, + "learning_rate": 9.460264308099805e-06, + "loss": 0.6957, + "step": 736 + }, + { + "epoch": 0.17518421678155455, + "grad_norm": 1.20964385354742, + "learning_rate": 9.458523055481658e-06, + "loss": 0.685, + "step": 737 + }, + { + "epoch": 0.17542191585452815, + "grad_norm": 1.3136946636916738, + "learning_rate": 9.456779159417686e-06, + "loss": 0.6518, + "step": 738 + }, + { + "epoch": 0.1756596149275018, + "grad_norm": 1.1709746732882924, + "learning_rate": 9.45503262094184e-06, + "loss": 0.677, + "step": 739 + }, + { + "epoch": 0.1758973140004754, + "grad_norm": 1.8045276744943528, + "learning_rate": 9.453283441089637e-06, + "loss": 0.6921, + "step": 740 + }, + { + "epoch": 0.176135013073449, + "grad_norm": 1.3579091677123727, + "learning_rate": 9.45153162089816e-06, + "loss": 0.6074, + "step": 741 + }, + { + "epoch": 0.17637271214642264, + "grad_norm": 1.7410957475746205, + "learning_rate": 9.44977716140606e-06, + "loss": 0.6956, + "step": 742 + }, + { + "epoch": 0.17661041121939625, + "grad_norm": 1.2953557113769454, + "learning_rate": 9.44802006365355e-06, + "loss": 0.6316, + "step": 743 + }, + { + "epoch": 0.17684811029236985, + "grad_norm": 1.3881278759568112, + "learning_rate": 9.446260328682406e-06, + "loss": 0.6802, + "step": 744 + }, + { + "epoch": 0.1770858093653435, + "grad_norm": 1.2520334546607532, + "learning_rate": 9.444497957535975e-06, + "loss": 0.6722, + "step": 745 + }, + { + "epoch": 0.1773235084383171, + "grad_norm": 1.166275847340381, + "learning_rate": 9.44273295125916e-06, + "loss": 0.6661, + "step": 746 + }, + { + "epoch": 0.1775612075112907, + "grad_norm": 1.4376358176049073, + "learning_rate": 9.440965310898425e-06, + "loss": 0.5988, + "step": 747 + }, + { + "epoch": 0.1777989065842643, + "grad_norm": 1.261544738163209, + "learning_rate": 9.439195037501803e-06, + "loss": 0.6277, + "step": 748 + }, + { + "epoch": 0.17803660565723795, + "grad_norm": 1.3220479019934397, + "learning_rate": 9.437422132118883e-06, + "loss": 0.68, + "step": 749 + }, + { + "epoch": 0.17827430473021155, + "grad_norm": 1.1689331284387954, + "learning_rate": 9.435646595800816e-06, + "loss": 0.6449, + "step": 750 + }, + { + "epoch": 0.17851200380318516, + "grad_norm": 1.4152634514608904, + "learning_rate": 9.43386842960031e-06, + "loss": 0.7092, + "step": 751 + }, + { + "epoch": 0.1787497028761588, + "grad_norm": 1.297396339433621, + "learning_rate": 9.432087634571638e-06, + "loss": 0.6645, + "step": 752 + }, + { + "epoch": 0.1789874019491324, + "grad_norm": 1.5113615846185133, + "learning_rate": 9.430304211770627e-06, + "loss": 0.6549, + "step": 753 + }, + { + "epoch": 0.179225101022106, + "grad_norm": 1.2811352465793735, + "learning_rate": 9.428518162254662e-06, + "loss": 0.6617, + "step": 754 + }, + { + "epoch": 0.17946280009507962, + "grad_norm": 1.1649618550534098, + "learning_rate": 9.426729487082691e-06, + "loss": 0.6872, + "step": 755 + }, + { + "epoch": 0.17970049916805325, + "grad_norm": 1.8503830226268907, + "learning_rate": 9.42493818731521e-06, + "loss": 0.6969, + "step": 756 + }, + { + "epoch": 0.17993819824102686, + "grad_norm": 1.1320920966650239, + "learning_rate": 9.423144264014278e-06, + "loss": 0.6549, + "step": 757 + }, + { + "epoch": 0.18017589731400047, + "grad_norm": 1.7239232422251958, + "learning_rate": 9.421347718243505e-06, + "loss": 0.7183, + "step": 758 + }, + { + "epoch": 0.1804135963869741, + "grad_norm": 1.3043503864031691, + "learning_rate": 9.419548551068061e-06, + "loss": 0.6165, + "step": 759 + }, + { + "epoch": 0.1806512954599477, + "grad_norm": 1.2402870713298517, + "learning_rate": 9.417746763554664e-06, + "loss": 0.6851, + "step": 760 + }, + { + "epoch": 0.18088899453292132, + "grad_norm": 1.2474760646550012, + "learning_rate": 9.415942356771593e-06, + "loss": 0.7242, + "step": 761 + }, + { + "epoch": 0.18112669360589495, + "grad_norm": 1.13108307565329, + "learning_rate": 9.41413533178867e-06, + "loss": 0.6267, + "step": 762 + }, + { + "epoch": 0.18136439267886856, + "grad_norm": 1.4655565115011262, + "learning_rate": 9.41232568967728e-06, + "loss": 0.6785, + "step": 763 + }, + { + "epoch": 0.18160209175184217, + "grad_norm": 1.128332309218577, + "learning_rate": 9.41051343151035e-06, + "loss": 0.6305, + "step": 764 + }, + { + "epoch": 0.18183979082481577, + "grad_norm": 1.2005551293874188, + "learning_rate": 9.408698558362369e-06, + "loss": 0.652, + "step": 765 + }, + { + "epoch": 0.1820774898977894, + "grad_norm": 1.6124893598145378, + "learning_rate": 9.406881071309365e-06, + "loss": 0.6908, + "step": 766 + }, + { + "epoch": 0.18231518897076301, + "grad_norm": 1.258254875608052, + "learning_rate": 9.405060971428924e-06, + "loss": 0.6203, + "step": 767 + }, + { + "epoch": 0.18255288804373662, + "grad_norm": 1.3099771889527478, + "learning_rate": 9.403238259800176e-06, + "loss": 0.6631, + "step": 768 + }, + { + "epoch": 0.18279058711671026, + "grad_norm": 1.5067994897705936, + "learning_rate": 9.401412937503802e-06, + "loss": 0.6688, + "step": 769 + }, + { + "epoch": 0.18302828618968386, + "grad_norm": 1.3071098860226338, + "learning_rate": 9.399585005622032e-06, + "loss": 0.6895, + "step": 770 + }, + { + "epoch": 0.18326598526265747, + "grad_norm": 1.327692464026782, + "learning_rate": 9.39775446523864e-06, + "loss": 0.6597, + "step": 771 + }, + { + "epoch": 0.18350368433563108, + "grad_norm": 1.2883108165054342, + "learning_rate": 9.39592131743895e-06, + "loss": 0.6764, + "step": 772 + }, + { + "epoch": 0.1837413834086047, + "grad_norm": 1.2501391049995367, + "learning_rate": 9.394085563309827e-06, + "loss": 0.6375, + "step": 773 + }, + { + "epoch": 0.18397908248157832, + "grad_norm": 1.8109057521092475, + "learning_rate": 9.392247203939689e-06, + "loss": 0.6744, + "step": 774 + }, + { + "epoch": 0.18421678155455193, + "grad_norm": 1.2387535903793223, + "learning_rate": 9.39040624041849e-06, + "loss": 0.6801, + "step": 775 + }, + { + "epoch": 0.18445448062752556, + "grad_norm": 1.3767130775661054, + "learning_rate": 9.388562673837735e-06, + "loss": 0.6805, + "step": 776 + }, + { + "epoch": 0.18469217970049917, + "grad_norm": 1.4673454430008814, + "learning_rate": 9.386716505290467e-06, + "loss": 0.654, + "step": 777 + }, + { + "epoch": 0.18492987877347278, + "grad_norm": 1.220202243297528, + "learning_rate": 9.384867735871278e-06, + "loss": 0.6571, + "step": 778 + }, + { + "epoch": 0.18516757784644639, + "grad_norm": 1.3768845526900229, + "learning_rate": 9.383016366676294e-06, + "loss": 0.6525, + "step": 779 + }, + { + "epoch": 0.18540527691942002, + "grad_norm": 1.176565745494909, + "learning_rate": 9.381162398803188e-06, + "loss": 0.6617, + "step": 780 + }, + { + "epoch": 0.18564297599239363, + "grad_norm": 1.3601475297348964, + "learning_rate": 9.379305833351174e-06, + "loss": 0.6675, + "step": 781 + }, + { + "epoch": 0.18588067506536723, + "grad_norm": 1.1375781814024721, + "learning_rate": 9.377446671421001e-06, + "loss": 0.666, + "step": 782 + }, + { + "epoch": 0.18611837413834087, + "grad_norm": 1.5585413109841573, + "learning_rate": 9.375584914114963e-06, + "loss": 0.6775, + "step": 783 + }, + { + "epoch": 0.18635607321131448, + "grad_norm": 1.37658470665187, + "learning_rate": 9.373720562536892e-06, + "loss": 0.6989, + "step": 784 + }, + { + "epoch": 0.18659377228428808, + "grad_norm": 1.2164835917174417, + "learning_rate": 9.371853617792158e-06, + "loss": 0.6362, + "step": 785 + }, + { + "epoch": 0.18683147135726172, + "grad_norm": 1.2567118768778565, + "learning_rate": 9.369984080987659e-06, + "loss": 0.682, + "step": 786 + }, + { + "epoch": 0.18706917043023533, + "grad_norm": 1.2836896080536042, + "learning_rate": 9.368111953231849e-06, + "loss": 0.6383, + "step": 787 + }, + { + "epoch": 0.18730686950320893, + "grad_norm": 1.1387623719101099, + "learning_rate": 9.3662372356347e-06, + "loss": 0.6779, + "step": 788 + }, + { + "epoch": 0.18754456857618254, + "grad_norm": 1.299147858974149, + "learning_rate": 9.36435992930773e-06, + "loss": 0.6839, + "step": 789 + }, + { + "epoch": 0.18778226764915618, + "grad_norm": 1.4777218388986602, + "learning_rate": 9.362480035363987e-06, + "loss": 0.7001, + "step": 790 + }, + { + "epoch": 0.18801996672212978, + "grad_norm": 1.2345993562434474, + "learning_rate": 9.360597554918055e-06, + "loss": 0.6766, + "step": 791 + }, + { + "epoch": 0.1882576657951034, + "grad_norm": 1.5645806974540597, + "learning_rate": 9.358712489086053e-06, + "loss": 0.6355, + "step": 792 + }, + { + "epoch": 0.18849536486807703, + "grad_norm": 2.4267308807101737, + "learning_rate": 9.35682483898563e-06, + "loss": 0.676, + "step": 793 + }, + { + "epoch": 0.18873306394105063, + "grad_norm": 1.2123979078685008, + "learning_rate": 9.354934605735968e-06, + "loss": 0.6546, + "step": 794 + }, + { + "epoch": 0.18897076301402424, + "grad_norm": 1.374821237471487, + "learning_rate": 9.353041790457783e-06, + "loss": 0.6958, + "step": 795 + }, + { + "epoch": 0.18920846208699785, + "grad_norm": 1.2918148194670203, + "learning_rate": 9.351146394273317e-06, + "loss": 0.6755, + "step": 796 + }, + { + "epoch": 0.18944616115997148, + "grad_norm": 1.8106874561127506, + "learning_rate": 9.349248418306347e-06, + "loss": 0.6941, + "step": 797 + }, + { + "epoch": 0.1896838602329451, + "grad_norm": 1.2531847146755244, + "learning_rate": 9.347347863682177e-06, + "loss": 0.6879, + "step": 798 + }, + { + "epoch": 0.1899215593059187, + "grad_norm": 1.4714454056246238, + "learning_rate": 9.345444731527642e-06, + "loss": 0.6985, + "step": 799 + }, + { + "epoch": 0.19015925837889233, + "grad_norm": 1.2930478577688163, + "learning_rate": 9.343539022971103e-06, + "loss": 0.6666, + "step": 800 + }, + { + "epoch": 0.19015925837889233, + "eval_loss": 0.6583006381988525, + "eval_runtime": 278.1783, + "eval_samples_per_second": 39.104, + "eval_steps_per_second": 1.222, + "step": 800 + }, + { + "epoch": 0.19039695745186594, + "grad_norm": 1.2709632808858886, + "learning_rate": 9.341630739142447e-06, + "loss": 0.6581, + "step": 801 + }, + { + "epoch": 0.19063465652483955, + "grad_norm": 1.2691924700560537, + "learning_rate": 9.339719881173093e-06, + "loss": 0.6462, + "step": 802 + }, + { + "epoch": 0.19087235559781318, + "grad_norm": 1.210169090263333, + "learning_rate": 9.33780645019598e-06, + "loss": 0.6918, + "step": 803 + }, + { + "epoch": 0.1911100546707868, + "grad_norm": 1.1849589491887267, + "learning_rate": 9.335890447345581e-06, + "loss": 0.6647, + "step": 804 + }, + { + "epoch": 0.1913477537437604, + "grad_norm": 1.1585012329869286, + "learning_rate": 9.333971873757885e-06, + "loss": 0.6619, + "step": 805 + }, + { + "epoch": 0.191585452816734, + "grad_norm": 1.1866933404114324, + "learning_rate": 9.33205073057041e-06, + "loss": 0.604, + "step": 806 + }, + { + "epoch": 0.19182315188970764, + "grad_norm": 1.4152554728822926, + "learning_rate": 9.330127018922195e-06, + "loss": 0.6907, + "step": 807 + }, + { + "epoch": 0.19206085096268125, + "grad_norm": 1.3856567391547028, + "learning_rate": 9.328200739953803e-06, + "loss": 0.7643, + "step": 808 + }, + { + "epoch": 0.19229855003565485, + "grad_norm": 1.613157140618603, + "learning_rate": 9.326271894807323e-06, + "loss": 0.6533, + "step": 809 + }, + { + "epoch": 0.1925362491086285, + "grad_norm": 1.2230974567141086, + "learning_rate": 9.32434048462636e-06, + "loss": 0.6607, + "step": 810 + }, + { + "epoch": 0.1927739481816021, + "grad_norm": 1.4108794200163257, + "learning_rate": 9.32240651055604e-06, + "loss": 0.624, + "step": 811 + }, + { + "epoch": 0.1930116472545757, + "grad_norm": 1.3723085363963423, + "learning_rate": 9.320469973743013e-06, + "loss": 0.667, + "step": 812 + }, + { + "epoch": 0.1932493463275493, + "grad_norm": 2.716123760939148, + "learning_rate": 9.318530875335445e-06, + "loss": 0.6939, + "step": 813 + }, + { + "epoch": 0.19348704540052294, + "grad_norm": 1.3681403542960693, + "learning_rate": 9.316589216483022e-06, + "loss": 0.624, + "step": 814 + }, + { + "epoch": 0.19372474447349655, + "grad_norm": 1.353648172764943, + "learning_rate": 9.31464499833695e-06, + "loss": 0.6179, + "step": 815 + }, + { + "epoch": 0.19396244354647016, + "grad_norm": 1.3101477414364895, + "learning_rate": 9.312698222049947e-06, + "loss": 0.6588, + "step": 816 + }, + { + "epoch": 0.1942001426194438, + "grad_norm": 1.146400949362108, + "learning_rate": 9.310748888776254e-06, + "loss": 0.6966, + "step": 817 + }, + { + "epoch": 0.1944378416924174, + "grad_norm": 1.1582291988811853, + "learning_rate": 9.308796999671625e-06, + "loss": 0.6582, + "step": 818 + }, + { + "epoch": 0.194675540765391, + "grad_norm": 1.4022099996140174, + "learning_rate": 9.306842555893329e-06, + "loss": 0.6408, + "step": 819 + }, + { + "epoch": 0.19491323983836464, + "grad_norm": 1.3010576921004269, + "learning_rate": 9.30488555860015e-06, + "loss": 0.6874, + "step": 820 + }, + { + "epoch": 0.19515093891133825, + "grad_norm": 1.255773653111657, + "learning_rate": 9.302926008952386e-06, + "loss": 0.6102, + "step": 821 + }, + { + "epoch": 0.19538863798431186, + "grad_norm": 1.2275702107106816, + "learning_rate": 9.30096390811185e-06, + "loss": 0.699, + "step": 822 + }, + { + "epoch": 0.19562633705728547, + "grad_norm": 1.4335950987502677, + "learning_rate": 9.298999257241862e-06, + "loss": 0.6883, + "step": 823 + }, + { + "epoch": 0.1958640361302591, + "grad_norm": 1.0720376830427616, + "learning_rate": 9.297032057507264e-06, + "loss": 0.6311, + "step": 824 + }, + { + "epoch": 0.1961017352032327, + "grad_norm": 1.3166117197096758, + "learning_rate": 9.2950623100744e-06, + "loss": 0.6348, + "step": 825 + }, + { + "epoch": 0.19633943427620631, + "grad_norm": 1.24959232074787, + "learning_rate": 9.293090016111128e-06, + "loss": 0.6174, + "step": 826 + }, + { + "epoch": 0.19657713334917995, + "grad_norm": 1.169370108829438, + "learning_rate": 9.291115176786814e-06, + "loss": 0.6505, + "step": 827 + }, + { + "epoch": 0.19681483242215356, + "grad_norm": 1.402147668928426, + "learning_rate": 9.289137793272339e-06, + "loss": 0.6963, + "step": 828 + }, + { + "epoch": 0.19705253149512716, + "grad_norm": 1.3402189760379581, + "learning_rate": 9.287157866740082e-06, + "loss": 0.6462, + "step": 829 + }, + { + "epoch": 0.19729023056810077, + "grad_norm": 1.4383961295100487, + "learning_rate": 9.28517539836394e-06, + "loss": 0.6449, + "step": 830 + }, + { + "epoch": 0.1975279296410744, + "grad_norm": 1.4428924291163667, + "learning_rate": 9.283190389319315e-06, + "loss": 0.6547, + "step": 831 + }, + { + "epoch": 0.19776562871404801, + "grad_norm": 1.2065674707477967, + "learning_rate": 9.281202840783109e-06, + "loss": 0.6393, + "step": 832 + }, + { + "epoch": 0.19800332778702162, + "grad_norm": 1.3015666716013714, + "learning_rate": 9.279212753933735e-06, + "loss": 0.6808, + "step": 833 + }, + { + "epoch": 0.19824102685999526, + "grad_norm": 1.2552839555329476, + "learning_rate": 9.27722012995111e-06, + "loss": 0.6986, + "step": 834 + }, + { + "epoch": 0.19847872593296886, + "grad_norm": 1.163185441178097, + "learning_rate": 9.275224970016656e-06, + "loss": 0.657, + "step": 835 + }, + { + "epoch": 0.19871642500594247, + "grad_norm": 1.3190372635120855, + "learning_rate": 9.273227275313298e-06, + "loss": 0.5945, + "step": 836 + }, + { + "epoch": 0.1989541240789161, + "grad_norm": 1.1969028354825575, + "learning_rate": 9.271227047025462e-06, + "loss": 0.5999, + "step": 837 + }, + { + "epoch": 0.1991918231518897, + "grad_norm": 1.4794513237188167, + "learning_rate": 9.26922428633908e-06, + "loss": 0.6307, + "step": 838 + }, + { + "epoch": 0.19942952222486332, + "grad_norm": 1.2005413873474187, + "learning_rate": 9.26721899444158e-06, + "loss": 0.6884, + "step": 839 + }, + { + "epoch": 0.19966722129783693, + "grad_norm": 1.5348922238423208, + "learning_rate": 9.265211172521898e-06, + "loss": 0.6151, + "step": 840 + }, + { + "epoch": 0.19990492037081056, + "grad_norm": 1.1990187350307733, + "learning_rate": 9.263200821770462e-06, + "loss": 0.6353, + "step": 841 + }, + { + "epoch": 0.20014261944378417, + "grad_norm": 1.7184146349090894, + "learning_rate": 9.261187943379206e-06, + "loss": 0.5485, + "step": 842 + }, + { + "epoch": 0.20038031851675778, + "grad_norm": 1.3604882062087447, + "learning_rate": 9.259172538541558e-06, + "loss": 0.679, + "step": 843 + }, + { + "epoch": 0.2006180175897314, + "grad_norm": 1.1783089932796746, + "learning_rate": 9.257154608452447e-06, + "loss": 0.6298, + "step": 844 + }, + { + "epoch": 0.20085571666270502, + "grad_norm": 1.326304511381228, + "learning_rate": 9.2551341543083e-06, + "loss": 0.6633, + "step": 845 + }, + { + "epoch": 0.20109341573567863, + "grad_norm": 1.1998852932675643, + "learning_rate": 9.253111177307035e-06, + "loss": 0.6577, + "step": 846 + }, + { + "epoch": 0.20133111480865223, + "grad_norm": 1.4493616879438864, + "learning_rate": 9.251085678648072e-06, + "loss": 0.6484, + "step": 847 + }, + { + "epoch": 0.20156881388162587, + "grad_norm": 1.2309282684693061, + "learning_rate": 9.249057659532321e-06, + "loss": 0.6606, + "step": 848 + }, + { + "epoch": 0.20180651295459948, + "grad_norm": 1.165667094406585, + "learning_rate": 9.247027121162192e-06, + "loss": 0.6563, + "step": 849 + }, + { + "epoch": 0.20204421202757308, + "grad_norm": 1.3647062078673995, + "learning_rate": 9.244994064741583e-06, + "loss": 0.6716, + "step": 850 + }, + { + "epoch": 0.20228191110054672, + "grad_norm": 1.8315432452420926, + "learning_rate": 9.242958491475888e-06, + "loss": 0.6586, + "step": 851 + }, + { + "epoch": 0.20251961017352033, + "grad_norm": 1.4479328593460203, + "learning_rate": 9.240920402571995e-06, + "loss": 0.6313, + "step": 852 + }, + { + "epoch": 0.20275730924649393, + "grad_norm": 1.317031615442439, + "learning_rate": 9.238879799238278e-06, + "loss": 0.6662, + "step": 853 + }, + { + "epoch": 0.20299500831946754, + "grad_norm": 1.3629903864871173, + "learning_rate": 9.236836682684604e-06, + "loss": 0.6864, + "step": 854 + }, + { + "epoch": 0.20323270739244118, + "grad_norm": 1.283432501921479, + "learning_rate": 9.234791054122336e-06, + "loss": 0.6442, + "step": 855 + }, + { + "epoch": 0.20347040646541478, + "grad_norm": 1.24398210376432, + "learning_rate": 9.23274291476432e-06, + "loss": 0.6289, + "step": 856 + }, + { + "epoch": 0.2037081055383884, + "grad_norm": 1.2827394887979142, + "learning_rate": 9.230692265824888e-06, + "loss": 0.6579, + "step": 857 + }, + { + "epoch": 0.20394580461136202, + "grad_norm": 2.0628506083412144, + "learning_rate": 9.228639108519867e-06, + "loss": 0.6361, + "step": 858 + }, + { + "epoch": 0.20418350368433563, + "grad_norm": 1.5154694102460888, + "learning_rate": 9.22658344406657e-06, + "loss": 0.6337, + "step": 859 + }, + { + "epoch": 0.20442120275730924, + "grad_norm": 1.43503438920679, + "learning_rate": 9.22452527368379e-06, + "loss": 0.629, + "step": 860 + }, + { + "epoch": 0.20465890183028287, + "grad_norm": 1.5097504100224672, + "learning_rate": 9.222464598591816e-06, + "loss": 0.6311, + "step": 861 + }, + { + "epoch": 0.20489660090325648, + "grad_norm": 1.1909906425880934, + "learning_rate": 9.220401420012412e-06, + "loss": 0.6266, + "step": 862 + }, + { + "epoch": 0.2051342999762301, + "grad_norm": 1.3540106960104932, + "learning_rate": 9.218335739168833e-06, + "loss": 0.6489, + "step": 863 + }, + { + "epoch": 0.2053719990492037, + "grad_norm": 1.2197139601313547, + "learning_rate": 9.216267557285814e-06, + "loss": 0.6375, + "step": 864 + }, + { + "epoch": 0.20560969812217733, + "grad_norm": 1.1435972583811393, + "learning_rate": 9.214196875589577e-06, + "loss": 0.5947, + "step": 865 + }, + { + "epoch": 0.20584739719515094, + "grad_norm": 1.371724974339111, + "learning_rate": 9.212123695307819e-06, + "loss": 0.6501, + "step": 866 + }, + { + "epoch": 0.20608509626812455, + "grad_norm": 1.1640356627687947, + "learning_rate": 9.210048017669727e-06, + "loss": 0.6712, + "step": 867 + }, + { + "epoch": 0.20632279534109818, + "grad_norm": 1.1762498776799661, + "learning_rate": 9.207969843905963e-06, + "loss": 0.6379, + "step": 868 + }, + { + "epoch": 0.2065604944140718, + "grad_norm": 1.3408124114750193, + "learning_rate": 9.205889175248668e-06, + "loss": 0.632, + "step": 869 + }, + { + "epoch": 0.2067981934870454, + "grad_norm": 1.8204607484247826, + "learning_rate": 9.203806012931469e-06, + "loss": 0.6809, + "step": 870 + }, + { + "epoch": 0.207035892560019, + "grad_norm": 1.628692571685787, + "learning_rate": 9.201720358189464e-06, + "loss": 0.6614, + "step": 871 + }, + { + "epoch": 0.20727359163299264, + "grad_norm": 1.9863938563139636, + "learning_rate": 9.199632212259233e-06, + "loss": 0.6544, + "step": 872 + }, + { + "epoch": 0.20751129070596624, + "grad_norm": 1.2051601246745332, + "learning_rate": 9.197541576378833e-06, + "loss": 0.6488, + "step": 873 + }, + { + "epoch": 0.20774898977893985, + "grad_norm": 1.4081307215373828, + "learning_rate": 9.195448451787792e-06, + "loss": 0.6454, + "step": 874 + }, + { + "epoch": 0.2079866888519135, + "grad_norm": 1.2405836418892948, + "learning_rate": 9.193352839727122e-06, + "loss": 0.6586, + "step": 875 + }, + { + "epoch": 0.2082243879248871, + "grad_norm": 1.2506506012820087, + "learning_rate": 9.191254741439303e-06, + "loss": 0.6412, + "step": 876 + }, + { + "epoch": 0.2084620869978607, + "grad_norm": 1.228615768965794, + "learning_rate": 9.189154158168293e-06, + "loss": 0.6761, + "step": 877 + }, + { + "epoch": 0.20869978607083434, + "grad_norm": 1.1682694057937573, + "learning_rate": 9.18705109115952e-06, + "loss": 0.6429, + "step": 878 + }, + { + "epoch": 0.20893748514380794, + "grad_norm": 1.2127750971831692, + "learning_rate": 9.18494554165989e-06, + "loss": 0.6486, + "step": 879 + }, + { + "epoch": 0.20917518421678155, + "grad_norm": 1.1967010711305721, + "learning_rate": 9.182837510917774e-06, + "loss": 0.6305, + "step": 880 + }, + { + "epoch": 0.20941288328975516, + "grad_norm": 1.3559495607773409, + "learning_rate": 9.18072700018302e-06, + "loss": 0.6881, + "step": 881 + }, + { + "epoch": 0.2096505823627288, + "grad_norm": 1.3599705163328788, + "learning_rate": 9.178614010706942e-06, + "loss": 0.6556, + "step": 882 + }, + { + "epoch": 0.2098882814357024, + "grad_norm": 1.2812554118138537, + "learning_rate": 9.176498543742328e-06, + "loss": 0.7021, + "step": 883 + }, + { + "epoch": 0.210125980508676, + "grad_norm": 1.4936353706991286, + "learning_rate": 9.17438060054343e-06, + "loss": 0.6212, + "step": 884 + }, + { + "epoch": 0.21036367958164964, + "grad_norm": 1.21739934862563, + "learning_rate": 9.17226018236597e-06, + "loss": 0.6402, + "step": 885 + }, + { + "epoch": 0.21060137865462325, + "grad_norm": 1.3643867414243598, + "learning_rate": 9.170137290467142e-06, + "loss": 0.582, + "step": 886 + }, + { + "epoch": 0.21083907772759686, + "grad_norm": 1.310835997867634, + "learning_rate": 9.168011926105598e-06, + "loss": 0.6374, + "step": 887 + }, + { + "epoch": 0.21107677680057046, + "grad_norm": 1.2376555299346104, + "learning_rate": 9.165884090541464e-06, + "loss": 0.6728, + "step": 888 + }, + { + "epoch": 0.2113144758735441, + "grad_norm": 1.2688367489792118, + "learning_rate": 9.163753785036324e-06, + "loss": 0.6307, + "step": 889 + }, + { + "epoch": 0.2115521749465177, + "grad_norm": 1.2628260206119686, + "learning_rate": 9.161621010853234e-06, + "loss": 0.6516, + "step": 890 + }, + { + "epoch": 0.21178987401949131, + "grad_norm": 1.2141908305486195, + "learning_rate": 9.15948576925671e-06, + "loss": 0.6613, + "step": 891 + }, + { + "epoch": 0.21202757309246495, + "grad_norm": 1.1772742695313945, + "learning_rate": 9.157348061512728e-06, + "loss": 0.6746, + "step": 892 + }, + { + "epoch": 0.21226527216543856, + "grad_norm": 2.331235078893807, + "learning_rate": 9.155207888888729e-06, + "loss": 0.6162, + "step": 893 + }, + { + "epoch": 0.21250297123841216, + "grad_norm": 1.2736316307870903, + "learning_rate": 9.153065252653617e-06, + "loss": 0.6233, + "step": 894 + }, + { + "epoch": 0.2127406703113858, + "grad_norm": 1.4179094331611075, + "learning_rate": 9.150920154077753e-06, + "loss": 0.6334, + "step": 895 + }, + { + "epoch": 0.2129783693843594, + "grad_norm": 1.2408494485332033, + "learning_rate": 9.148772594432965e-06, + "loss": 0.6643, + "step": 896 + }, + { + "epoch": 0.213216068457333, + "grad_norm": 1.6942051648297631, + "learning_rate": 9.146622574992528e-06, + "loss": 0.677, + "step": 897 + }, + { + "epoch": 0.21345376753030662, + "grad_norm": 1.0925023339811777, + "learning_rate": 9.144470097031187e-06, + "loss": 0.6706, + "step": 898 + }, + { + "epoch": 0.21369146660328026, + "grad_norm": 1.1473000608166426, + "learning_rate": 9.14231516182514e-06, + "loss": 0.6268, + "step": 899 + }, + { + "epoch": 0.21392916567625386, + "grad_norm": 1.1229160397674556, + "learning_rate": 9.140157770652041e-06, + "loss": 0.6393, + "step": 900 + }, + { + "epoch": 0.21392916567625386, + "eval_loss": 0.6523029208183289, + "eval_runtime": 278.0536, + "eval_samples_per_second": 39.122, + "eval_steps_per_second": 1.223, + "step": 900 + }, + { + "epoch": 0.21416686474922747, + "grad_norm": 1.2215340933561183, + "learning_rate": 9.137997924791e-06, + "loss": 0.6493, + "step": 901 + }, + { + "epoch": 0.2144045638222011, + "grad_norm": 1.385981395588334, + "learning_rate": 9.135835625522585e-06, + "loss": 0.6688, + "step": 902 + }, + { + "epoch": 0.2146422628951747, + "grad_norm": 1.257071859882414, + "learning_rate": 9.133670874128818e-06, + "loss": 0.6655, + "step": 903 + }, + { + "epoch": 0.21487996196814832, + "grad_norm": 1.172647620386631, + "learning_rate": 9.131503671893173e-06, + "loss": 0.6183, + "step": 904 + }, + { + "epoch": 0.21511766104112193, + "grad_norm": 1.185796558462135, + "learning_rate": 9.129334020100577e-06, + "loss": 0.6729, + "step": 905 + }, + { + "epoch": 0.21535536011409556, + "grad_norm": 1.7430934884807652, + "learning_rate": 9.127161920037414e-06, + "loss": 0.7307, + "step": 906 + }, + { + "epoch": 0.21559305918706917, + "grad_norm": 1.2653134473128114, + "learning_rate": 9.124987372991512e-06, + "loss": 0.7062, + "step": 907 + }, + { + "epoch": 0.21583075826004278, + "grad_norm": 1.3775541238427094, + "learning_rate": 9.122810380252156e-06, + "loss": 0.6597, + "step": 908 + }, + { + "epoch": 0.2160684573330164, + "grad_norm": 1.2758738393171105, + "learning_rate": 9.120630943110078e-06, + "loss": 0.6178, + "step": 909 + }, + { + "epoch": 0.21630615640599002, + "grad_norm": 1.1327888110810036, + "learning_rate": 9.118449062857463e-06, + "loss": 0.6409, + "step": 910 + }, + { + "epoch": 0.21654385547896363, + "grad_norm": 1.3868108760116626, + "learning_rate": 9.116264740787937e-06, + "loss": 0.6395, + "step": 911 + }, + { + "epoch": 0.21678155455193726, + "grad_norm": 1.293355706631399, + "learning_rate": 9.11407797819658e-06, + "loss": 0.6385, + "step": 912 + }, + { + "epoch": 0.21701925362491087, + "grad_norm": 1.2487869418623918, + "learning_rate": 9.11188877637992e-06, + "loss": 0.7105, + "step": 913 + }, + { + "epoch": 0.21725695269788448, + "grad_norm": 1.5163974333961359, + "learning_rate": 9.109697136635927e-06, + "loss": 0.6664, + "step": 914 + }, + { + "epoch": 0.21749465177085808, + "grad_norm": 1.1922689341811865, + "learning_rate": 9.107503060264018e-06, + "loss": 0.6551, + "step": 915 + }, + { + "epoch": 0.21773235084383172, + "grad_norm": 2.9046836309569684, + "learning_rate": 9.105306548565054e-06, + "loss": 0.6447, + "step": 916 + }, + { + "epoch": 0.21797004991680533, + "grad_norm": 1.1935851824512065, + "learning_rate": 9.103107602841341e-06, + "loss": 0.6697, + "step": 917 + }, + { + "epoch": 0.21820774898977893, + "grad_norm": 1.1889622842712737, + "learning_rate": 9.100906224396628e-06, + "loss": 0.6656, + "step": 918 + }, + { + "epoch": 0.21844544806275257, + "grad_norm": 1.2061772151251782, + "learning_rate": 9.098702414536107e-06, + "loss": 0.6779, + "step": 919 + }, + { + "epoch": 0.21868314713572617, + "grad_norm": 1.3844149291604109, + "learning_rate": 9.096496174566408e-06, + "loss": 0.6607, + "step": 920 + }, + { + "epoch": 0.21892084620869978, + "grad_norm": 1.6512970091714194, + "learning_rate": 9.094287505795607e-06, + "loss": 0.6403, + "step": 921 + }, + { + "epoch": 0.2191585452816734, + "grad_norm": 1.4280155687021145, + "learning_rate": 9.092076409533218e-06, + "loss": 0.6128, + "step": 922 + }, + { + "epoch": 0.21939624435464702, + "grad_norm": 1.240832534523375, + "learning_rate": 9.089862887090191e-06, + "loss": 0.6678, + "step": 923 + }, + { + "epoch": 0.21963394342762063, + "grad_norm": 1.486312055387504, + "learning_rate": 9.08764693977892e-06, + "loss": 0.647, + "step": 924 + }, + { + "epoch": 0.21987164250059424, + "grad_norm": 1.3832073210540634, + "learning_rate": 9.085428568913233e-06, + "loss": 0.7326, + "step": 925 + }, + { + "epoch": 0.22010934157356787, + "grad_norm": 1.2872689965952084, + "learning_rate": 9.083207775808395e-06, + "loss": 0.6362, + "step": 926 + }, + { + "epoch": 0.22034704064654148, + "grad_norm": 1.334145303778825, + "learning_rate": 9.08098456178111e-06, + "loss": 0.6594, + "step": 927 + }, + { + "epoch": 0.2205847397195151, + "grad_norm": 1.2578395882986613, + "learning_rate": 9.078758928149512e-06, + "loss": 0.6842, + "step": 928 + }, + { + "epoch": 0.2208224387924887, + "grad_norm": 1.137737608477493, + "learning_rate": 9.076530876233175e-06, + "loss": 0.6556, + "step": 929 + }, + { + "epoch": 0.22106013786546233, + "grad_norm": 1.0768103697235134, + "learning_rate": 9.074300407353107e-06, + "loss": 0.704, + "step": 930 + }, + { + "epoch": 0.22129783693843594, + "grad_norm": 1.3086318499269856, + "learning_rate": 9.072067522831743e-06, + "loss": 0.643, + "step": 931 + }, + { + "epoch": 0.22153553601140955, + "grad_norm": 1.1415042049874216, + "learning_rate": 9.069832223992955e-06, + "loss": 0.6259, + "step": 932 + }, + { + "epoch": 0.22177323508438318, + "grad_norm": 1.1257099524123508, + "learning_rate": 9.067594512162047e-06, + "loss": 0.6332, + "step": 933 + }, + { + "epoch": 0.2220109341573568, + "grad_norm": 1.3162761467394548, + "learning_rate": 9.06535438866575e-06, + "loss": 0.6242, + "step": 934 + }, + { + "epoch": 0.2222486332303304, + "grad_norm": 1.5664263387041373, + "learning_rate": 9.06311185483223e-06, + "loss": 0.6397, + "step": 935 + }, + { + "epoch": 0.22248633230330403, + "grad_norm": 1.3404454388100735, + "learning_rate": 9.060866911991074e-06, + "loss": 0.6754, + "step": 936 + }, + { + "epoch": 0.22272403137627764, + "grad_norm": 1.2818778031499154, + "learning_rate": 9.058619561473308e-06, + "loss": 0.7013, + "step": 937 + }, + { + "epoch": 0.22296173044925124, + "grad_norm": 1.2038986139119752, + "learning_rate": 9.056369804611374e-06, + "loss": 0.6493, + "step": 938 + }, + { + "epoch": 0.22319942952222485, + "grad_norm": 1.5857882746207268, + "learning_rate": 9.054117642739151e-06, + "loss": 0.6261, + "step": 939 + }, + { + "epoch": 0.2234371285951985, + "grad_norm": 1.144864488352514, + "learning_rate": 9.051863077191938e-06, + "loss": 0.6444, + "step": 940 + }, + { + "epoch": 0.2236748276681721, + "grad_norm": 1.5130499945226639, + "learning_rate": 9.049606109306461e-06, + "loss": 0.6286, + "step": 941 + }, + { + "epoch": 0.2239125267411457, + "grad_norm": 1.2088411749296981, + "learning_rate": 9.04734674042087e-06, + "loss": 0.6345, + "step": 942 + }, + { + "epoch": 0.22415022581411934, + "grad_norm": 1.1548833178259146, + "learning_rate": 9.045084971874738e-06, + "loss": 0.5847, + "step": 943 + }, + { + "epoch": 0.22438792488709294, + "grad_norm": 1.2540733755424316, + "learning_rate": 9.04282080500906e-06, + "loss": 0.6517, + "step": 944 + }, + { + "epoch": 0.22462562396006655, + "grad_norm": 1.31588808058698, + "learning_rate": 9.040554241166257e-06, + "loss": 0.7059, + "step": 945 + }, + { + "epoch": 0.22486332303304016, + "grad_norm": 1.521207131639247, + "learning_rate": 9.038285281690166e-06, + "loss": 0.6404, + "step": 946 + }, + { + "epoch": 0.2251010221060138, + "grad_norm": 1.0848748008983908, + "learning_rate": 9.036013927926049e-06, + "loss": 0.6619, + "step": 947 + }, + { + "epoch": 0.2253387211789874, + "grad_norm": 1.2962662567934486, + "learning_rate": 9.033740181220582e-06, + "loss": 0.6677, + "step": 948 + }, + { + "epoch": 0.225576420251961, + "grad_norm": 1.3249070998854044, + "learning_rate": 9.031464042921866e-06, + "loss": 0.6896, + "step": 949 + }, + { + "epoch": 0.22581411932493464, + "grad_norm": 1.0988250850851342, + "learning_rate": 9.029185514379414e-06, + "loss": 0.5693, + "step": 950 + }, + { + "epoch": 0.22605181839790825, + "grad_norm": 1.2717351607678913, + "learning_rate": 9.026904596944163e-06, + "loss": 0.6829, + "step": 951 + }, + { + "epoch": 0.22628951747088186, + "grad_norm": 1.1486572152621932, + "learning_rate": 9.024621291968461e-06, + "loss": 0.6704, + "step": 952 + }, + { + "epoch": 0.2265272165438555, + "grad_norm": 1.2364522956686, + "learning_rate": 9.022335600806073e-06, + "loss": 0.6635, + "step": 953 + }, + { + "epoch": 0.2267649156168291, + "grad_norm": 1.5164266635462873, + "learning_rate": 9.020047524812177e-06, + "loss": 0.6178, + "step": 954 + }, + { + "epoch": 0.2270026146898027, + "grad_norm": 1.176691653216935, + "learning_rate": 9.017757065343368e-06, + "loss": 0.6654, + "step": 955 + }, + { + "epoch": 0.2272403137627763, + "grad_norm": 1.2017488529727716, + "learning_rate": 9.015464223757656e-06, + "loss": 0.5986, + "step": 956 + }, + { + "epoch": 0.22747801283574995, + "grad_norm": 1.136333459431739, + "learning_rate": 9.013169001414458e-06, + "loss": 0.6173, + "step": 957 + }, + { + "epoch": 0.22771571190872356, + "grad_norm": 1.3596324636056099, + "learning_rate": 9.010871399674603e-06, + "loss": 0.6859, + "step": 958 + }, + { + "epoch": 0.22795341098169716, + "grad_norm": 1.253313366935477, + "learning_rate": 9.008571419900334e-06, + "loss": 0.6393, + "step": 959 + }, + { + "epoch": 0.2281911100546708, + "grad_norm": 1.275327400230184, + "learning_rate": 9.006269063455305e-06, + "loss": 0.6621, + "step": 960 + }, + { + "epoch": 0.2284288091276444, + "grad_norm": 1.2331836761583428, + "learning_rate": 9.003964331704574e-06, + "loss": 0.6732, + "step": 961 + }, + { + "epoch": 0.228666508200618, + "grad_norm": 1.3625822894824966, + "learning_rate": 9.001657226014609e-06, + "loss": 0.653, + "step": 962 + }, + { + "epoch": 0.22890420727359162, + "grad_norm": 1.253178290578461, + "learning_rate": 8.999347747753288e-06, + "loss": 0.6147, + "step": 963 + }, + { + "epoch": 0.22914190634656526, + "grad_norm": 1.3170082435038148, + "learning_rate": 8.997035898289897e-06, + "loss": 0.6486, + "step": 964 + }, + { + "epoch": 0.22937960541953886, + "grad_norm": 1.3428947467319914, + "learning_rate": 8.99472167899512e-06, + "loss": 0.5844, + "step": 965 + }, + { + "epoch": 0.22961730449251247, + "grad_norm": 2.3790797335203218, + "learning_rate": 8.992405091241055e-06, + "loss": 0.6875, + "step": 966 + }, + { + "epoch": 0.2298550035654861, + "grad_norm": 1.1806173567810558, + "learning_rate": 8.990086136401199e-06, + "loss": 0.5642, + "step": 967 + }, + { + "epoch": 0.2300927026384597, + "grad_norm": 3.859703711119646, + "learning_rate": 8.987764815850453e-06, + "loss": 0.6394, + "step": 968 + }, + { + "epoch": 0.23033040171143332, + "grad_norm": 1.2540884483010746, + "learning_rate": 8.985441130965123e-06, + "loss": 0.6803, + "step": 969 + }, + { + "epoch": 0.23056810078440695, + "grad_norm": 1.4378836894211577, + "learning_rate": 8.983115083122912e-06, + "loss": 0.7041, + "step": 970 + }, + { + "epoch": 0.23080579985738056, + "grad_norm": 1.4330355061439808, + "learning_rate": 8.980786673702932e-06, + "loss": 0.6552, + "step": 971 + }, + { + "epoch": 0.23104349893035417, + "grad_norm": 1.1674880263034924, + "learning_rate": 8.978455904085687e-06, + "loss": 0.6806, + "step": 972 + }, + { + "epoch": 0.23128119800332778, + "grad_norm": 1.2332528196127839, + "learning_rate": 8.976122775653087e-06, + "loss": 0.5984, + "step": 973 + }, + { + "epoch": 0.2315188970763014, + "grad_norm": 1.2022211800518827, + "learning_rate": 8.973787289788433e-06, + "loss": 0.6894, + "step": 974 + }, + { + "epoch": 0.23175659614927502, + "grad_norm": 1.1353491430142426, + "learning_rate": 8.97144944787643e-06, + "loss": 0.6755, + "step": 975 + }, + { + "epoch": 0.23199429522224863, + "grad_norm": 1.130800642775017, + "learning_rate": 8.969109251303181e-06, + "loss": 0.6317, + "step": 976 + }, + { + "epoch": 0.23223199429522226, + "grad_norm": 1.6187536312171678, + "learning_rate": 8.966766701456177e-06, + "loss": 0.659, + "step": 977 + }, + { + "epoch": 0.23246969336819587, + "grad_norm": 1.4179952222594545, + "learning_rate": 8.964421799724312e-06, + "loss": 0.6701, + "step": 978 + }, + { + "epoch": 0.23270739244116947, + "grad_norm": 1.5840796192454847, + "learning_rate": 8.96207454749787e-06, + "loss": 0.6333, + "step": 979 + }, + { + "epoch": 0.23294509151414308, + "grad_norm": 1.1762386071472137, + "learning_rate": 8.959724946168532e-06, + "loss": 0.6165, + "step": 980 + }, + { + "epoch": 0.23318279058711672, + "grad_norm": 1.304846509546937, + "learning_rate": 8.95737299712937e-06, + "loss": 0.6527, + "step": 981 + }, + { + "epoch": 0.23342048966009032, + "grad_norm": 1.5085537502573056, + "learning_rate": 8.955018701774848e-06, + "loss": 0.6157, + "step": 982 + }, + { + "epoch": 0.23365818873306393, + "grad_norm": 1.2300612083281444, + "learning_rate": 8.952662061500817e-06, + "loss": 0.6236, + "step": 983 + }, + { + "epoch": 0.23389588780603757, + "grad_norm": 1.8841971326631577, + "learning_rate": 8.950303077704529e-06, + "loss": 0.6385, + "step": 984 + }, + { + "epoch": 0.23413358687901117, + "grad_norm": 1.287571882316221, + "learning_rate": 8.947941751784614e-06, + "loss": 0.6716, + "step": 985 + }, + { + "epoch": 0.23437128595198478, + "grad_norm": 1.307367111235092, + "learning_rate": 8.945578085141098e-06, + "loss": 0.6557, + "step": 986 + }, + { + "epoch": 0.23460898502495842, + "grad_norm": 1.4762511894107844, + "learning_rate": 8.943212079175392e-06, + "loss": 0.6751, + "step": 987 + }, + { + "epoch": 0.23484668409793202, + "grad_norm": 1.7576302029761746, + "learning_rate": 8.940843735290294e-06, + "loss": 0.6404, + "step": 988 + }, + { + "epoch": 0.23508438317090563, + "grad_norm": 1.2698246671867257, + "learning_rate": 8.938473054889988e-06, + "loss": 0.6907, + "step": 989 + }, + { + "epoch": 0.23532208224387924, + "grad_norm": 1.4382340134552578, + "learning_rate": 8.936100039380045e-06, + "loss": 0.6634, + "step": 990 + }, + { + "epoch": 0.23555978131685287, + "grad_norm": 1.1228279325981758, + "learning_rate": 8.933724690167417e-06, + "loss": 0.6151, + "step": 991 + }, + { + "epoch": 0.23579748038982648, + "grad_norm": 1.2702950439793108, + "learning_rate": 8.931347008660443e-06, + "loss": 0.6764, + "step": 992 + }, + { + "epoch": 0.2360351794628001, + "grad_norm": 1.219995033472717, + "learning_rate": 8.928966996268847e-06, + "loss": 0.6357, + "step": 993 + }, + { + "epoch": 0.23627287853577372, + "grad_norm": 1.1673317153142275, + "learning_rate": 8.926584654403725e-06, + "loss": 0.6584, + "step": 994 + }, + { + "epoch": 0.23651057760874733, + "grad_norm": 1.3570648762974182, + "learning_rate": 8.924199984477566e-06, + "loss": 0.6976, + "step": 995 + }, + { + "epoch": 0.23674827668172094, + "grad_norm": 1.078482167299246, + "learning_rate": 8.921812987904233e-06, + "loss": 0.6373, + "step": 996 + }, + { + "epoch": 0.23698597575469454, + "grad_norm": 1.3404071741700359, + "learning_rate": 8.91942366609897e-06, + "loss": 0.6444, + "step": 997 + }, + { + "epoch": 0.23722367482766818, + "grad_norm": 1.3135273426256138, + "learning_rate": 8.917032020478396e-06, + "loss": 0.6274, + "step": 998 + }, + { + "epoch": 0.2374613739006418, + "grad_norm": 1.2248268009264456, + "learning_rate": 8.914638052460515e-06, + "loss": 0.6133, + "step": 999 + }, + { + "epoch": 0.2376990729736154, + "grad_norm": 1.1477802380014233, + "learning_rate": 8.912241763464702e-06, + "loss": 0.6255, + "step": 1000 + }, + { + "epoch": 0.2376990729736154, + "eval_loss": 0.6477440595626831, + "eval_runtime": 278.3468, + "eval_samples_per_second": 39.081, + "eval_steps_per_second": 1.221, + "step": 1000 + }, + { + "epoch": 0.23793677204658903, + "grad_norm": 1.4676651432293286, + "learning_rate": 8.909843154911711e-06, + "loss": 0.674, + "step": 1001 + }, + { + "epoch": 0.23817447111956264, + "grad_norm": 1.1931020001970898, + "learning_rate": 8.907442228223668e-06, + "loss": 0.6269, + "step": 1002 + }, + { + "epoch": 0.23841217019253624, + "grad_norm": 1.2695449452163898, + "learning_rate": 8.905038984824079e-06, + "loss": 0.6603, + "step": 1003 + }, + { + "epoch": 0.23864986926550985, + "grad_norm": 1.189694951268138, + "learning_rate": 8.902633426137817e-06, + "loss": 0.6696, + "step": 1004 + }, + { + "epoch": 0.23888756833848349, + "grad_norm": 1.3004320349077945, + "learning_rate": 8.900225553591132e-06, + "loss": 0.5958, + "step": 1005 + }, + { + "epoch": 0.2391252674114571, + "grad_norm": 1.1783049753152042, + "learning_rate": 8.897815368611648e-06, + "loss": 0.6564, + "step": 1006 + }, + { + "epoch": 0.2393629664844307, + "grad_norm": 1.6179598725005921, + "learning_rate": 8.895402872628352e-06, + "loss": 0.687, + "step": 1007 + }, + { + "epoch": 0.23960066555740434, + "grad_norm": 1.3046996492998664, + "learning_rate": 8.892988067071611e-06, + "loss": 0.6465, + "step": 1008 + }, + { + "epoch": 0.23983836463037794, + "grad_norm": 1.2421889717859609, + "learning_rate": 8.890570953373152e-06, + "loss": 0.6376, + "step": 1009 + }, + { + "epoch": 0.24007606370335155, + "grad_norm": 1.5779403013116822, + "learning_rate": 8.888151532966078e-06, + "loss": 0.6148, + "step": 1010 + }, + { + "epoch": 0.24031376277632518, + "grad_norm": 1.2688623220512247, + "learning_rate": 8.885729807284855e-06, + "loss": 0.6286, + "step": 1011 + }, + { + "epoch": 0.2405514618492988, + "grad_norm": 1.6408473779585402, + "learning_rate": 8.883305777765319e-06, + "loss": 0.6469, + "step": 1012 + }, + { + "epoch": 0.2407891609222724, + "grad_norm": 1.548129451906955, + "learning_rate": 8.880879445844668e-06, + "loss": 0.5831, + "step": 1013 + }, + { + "epoch": 0.241026859995246, + "grad_norm": 1.3438852793197966, + "learning_rate": 8.878450812961468e-06, + "loss": 0.6238, + "step": 1014 + }, + { + "epoch": 0.24126455906821964, + "grad_norm": 1.249026449186817, + "learning_rate": 8.87601988055565e-06, + "loss": 0.6498, + "step": 1015 + }, + { + "epoch": 0.24150225814119325, + "grad_norm": 1.2663053506867177, + "learning_rate": 8.873586650068506e-06, + "loss": 0.6871, + "step": 1016 + }, + { + "epoch": 0.24173995721416686, + "grad_norm": 1.258507635695742, + "learning_rate": 8.871151122942692e-06, + "loss": 0.659, + "step": 1017 + }, + { + "epoch": 0.2419776562871405, + "grad_norm": 1.21922287884461, + "learning_rate": 8.868713300622224e-06, + "loss": 0.6778, + "step": 1018 + }, + { + "epoch": 0.2422153553601141, + "grad_norm": 1.216718482576782, + "learning_rate": 8.866273184552483e-06, + "loss": 0.6197, + "step": 1019 + }, + { + "epoch": 0.2424530544330877, + "grad_norm": 1.1536536117197922, + "learning_rate": 8.863830776180203e-06, + "loss": 0.6139, + "step": 1020 + }, + { + "epoch": 0.2426907535060613, + "grad_norm": 1.3085895565809667, + "learning_rate": 8.861386076953485e-06, + "loss": 0.6813, + "step": 1021 + }, + { + "epoch": 0.24292845257903495, + "grad_norm": 1.1897067842495666, + "learning_rate": 8.858939088321781e-06, + "loss": 0.5984, + "step": 1022 + }, + { + "epoch": 0.24316615165200856, + "grad_norm": 1.4128110990615435, + "learning_rate": 8.856489811735904e-06, + "loss": 0.666, + "step": 1023 + }, + { + "epoch": 0.24340385072498216, + "grad_norm": 1.130838818885665, + "learning_rate": 8.854038248648025e-06, + "loss": 0.6546, + "step": 1024 + }, + { + "epoch": 0.2436415497979558, + "grad_norm": 1.5346741957125316, + "learning_rate": 8.851584400511668e-06, + "loss": 0.6081, + "step": 1025 + }, + { + "epoch": 0.2438792488709294, + "grad_norm": 1.2884727525736877, + "learning_rate": 8.849128268781714e-06, + "loss": 0.6152, + "step": 1026 + }, + { + "epoch": 0.244116947943903, + "grad_norm": 1.2510680346816143, + "learning_rate": 8.846669854914395e-06, + "loss": 0.6415, + "step": 1027 + }, + { + "epoch": 0.24435464701687665, + "grad_norm": 1.1713609580962958, + "learning_rate": 8.844209160367298e-06, + "loss": 0.6387, + "step": 1028 + }, + { + "epoch": 0.24459234608985025, + "grad_norm": 1.654502868137218, + "learning_rate": 8.841746186599362e-06, + "loss": 0.6312, + "step": 1029 + }, + { + "epoch": 0.24483004516282386, + "grad_norm": 1.1969465722055663, + "learning_rate": 8.839280935070877e-06, + "loss": 0.6206, + "step": 1030 + }, + { + "epoch": 0.24506774423579747, + "grad_norm": 1.213554547431693, + "learning_rate": 8.836813407243485e-06, + "loss": 0.6549, + "step": 1031 + }, + { + "epoch": 0.2453054433087711, + "grad_norm": 1.1462131371143032, + "learning_rate": 8.834343604580175e-06, + "loss": 0.6647, + "step": 1032 + }, + { + "epoch": 0.2455431423817447, + "grad_norm": 1.1819845069171273, + "learning_rate": 8.831871528545286e-06, + "loss": 0.5951, + "step": 1033 + }, + { + "epoch": 0.24578084145471832, + "grad_norm": 1.179441576395199, + "learning_rate": 8.829397180604506e-06, + "loss": 0.6377, + "step": 1034 + }, + { + "epoch": 0.24601854052769195, + "grad_norm": 1.1783482261936433, + "learning_rate": 8.826920562224868e-06, + "loss": 0.6568, + "step": 1035 + }, + { + "epoch": 0.24625623960066556, + "grad_norm": 1.1849971382323388, + "learning_rate": 8.824441674874753e-06, + "loss": 0.6661, + "step": 1036 + }, + { + "epoch": 0.24649393867363917, + "grad_norm": 1.7307631577122913, + "learning_rate": 8.821960520023884e-06, + "loss": 0.6935, + "step": 1037 + }, + { + "epoch": 0.24673163774661278, + "grad_norm": 1.3404024145849653, + "learning_rate": 8.819477099143334e-06, + "loss": 0.6762, + "step": 1038 + }, + { + "epoch": 0.2469693368195864, + "grad_norm": 1.1440415885157713, + "learning_rate": 8.816991413705515e-06, + "loss": 0.6681, + "step": 1039 + }, + { + "epoch": 0.24720703589256002, + "grad_norm": 1.200731028546892, + "learning_rate": 8.814503465184182e-06, + "loss": 0.6877, + "step": 1040 + }, + { + "epoch": 0.24744473496553362, + "grad_norm": 1.1717224808155042, + "learning_rate": 8.812013255054435e-06, + "loss": 0.6134, + "step": 1041 + }, + { + "epoch": 0.24768243403850726, + "grad_norm": 1.3078988820383377, + "learning_rate": 8.80952078479271e-06, + "loss": 0.6433, + "step": 1042 + }, + { + "epoch": 0.24792013311148087, + "grad_norm": 1.2109583198753988, + "learning_rate": 8.807026055876787e-06, + "loss": 0.6483, + "step": 1043 + }, + { + "epoch": 0.24815783218445447, + "grad_norm": 1.2648527300040195, + "learning_rate": 8.804529069785784e-06, + "loss": 0.624, + "step": 1044 + }, + { + "epoch": 0.2483955312574281, + "grad_norm": 1.1619529969796825, + "learning_rate": 8.802029828000157e-06, + "loss": 0.5952, + "step": 1045 + }, + { + "epoch": 0.24863323033040172, + "grad_norm": 1.1601259899618928, + "learning_rate": 8.799528332001697e-06, + "loss": 0.6699, + "step": 1046 + }, + { + "epoch": 0.24887092940337532, + "grad_norm": 1.1690521587541718, + "learning_rate": 8.797024583273536e-06, + "loss": 0.5842, + "step": 1047 + }, + { + "epoch": 0.24910862847634893, + "grad_norm": 1.358543734612153, + "learning_rate": 8.794518583300141e-06, + "loss": 0.6129, + "step": 1048 + }, + { + "epoch": 0.24934632754932257, + "grad_norm": 1.3809805784907774, + "learning_rate": 8.792010333567311e-06, + "loss": 0.635, + "step": 1049 + }, + { + "epoch": 0.24958402662229617, + "grad_norm": 1.3317747423482778, + "learning_rate": 8.789499835562178e-06, + "loss": 0.6265, + "step": 1050 + }, + { + "epoch": 0.24982172569526978, + "grad_norm": 1.3613814704394838, + "learning_rate": 8.786987090773214e-06, + "loss": 0.6593, + "step": 1051 + }, + { + "epoch": 0.2500594247682434, + "grad_norm": 1.4703366434178782, + "learning_rate": 8.784472100690215e-06, + "loss": 0.7052, + "step": 1052 + }, + { + "epoch": 0.250297123841217, + "grad_norm": 1.2309851839220152, + "learning_rate": 8.781954866804313e-06, + "loss": 0.6395, + "step": 1053 + }, + { + "epoch": 0.25053482291419066, + "grad_norm": 1.2249438230312817, + "learning_rate": 8.779435390607968e-06, + "loss": 0.6321, + "step": 1054 + }, + { + "epoch": 0.25077252198716427, + "grad_norm": 1.350227152207108, + "learning_rate": 8.776913673594968e-06, + "loss": 0.5857, + "step": 1055 + }, + { + "epoch": 0.2510102210601379, + "grad_norm": 1.1683927084289312, + "learning_rate": 8.774389717260439e-06, + "loss": 0.6298, + "step": 1056 + }, + { + "epoch": 0.2512479201331115, + "grad_norm": 1.134208088983512, + "learning_rate": 8.771863523100821e-06, + "loss": 0.6618, + "step": 1057 + }, + { + "epoch": 0.2514856192060851, + "grad_norm": 1.2263640127571174, + "learning_rate": 8.76933509261389e-06, + "loss": 0.6253, + "step": 1058 + }, + { + "epoch": 0.2517233182790587, + "grad_norm": 1.129506430342816, + "learning_rate": 8.766804427298745e-06, + "loss": 0.6444, + "step": 1059 + }, + { + "epoch": 0.2519610173520323, + "grad_norm": 1.3523002210471045, + "learning_rate": 8.76427152865581e-06, + "loss": 0.6373, + "step": 1060 + }, + { + "epoch": 0.25219871642500596, + "grad_norm": 1.2222011090212135, + "learning_rate": 8.761736398186834e-06, + "loss": 0.6379, + "step": 1061 + }, + { + "epoch": 0.25243641549797957, + "grad_norm": 1.2497661895834127, + "learning_rate": 8.759199037394888e-06, + "loss": 0.6327, + "step": 1062 + }, + { + "epoch": 0.2526741145709532, + "grad_norm": 1.268173498015707, + "learning_rate": 8.756659447784367e-06, + "loss": 0.6392, + "step": 1063 + }, + { + "epoch": 0.2529118136439268, + "grad_norm": 3.0350424315708957, + "learning_rate": 8.754117630860987e-06, + "loss": 0.6987, + "step": 1064 + }, + { + "epoch": 0.2531495127169004, + "grad_norm": 1.2101163493986367, + "learning_rate": 8.751573588131783e-06, + "loss": 0.6336, + "step": 1065 + }, + { + "epoch": 0.253387211789874, + "grad_norm": 1.225211240446433, + "learning_rate": 8.749027321105112e-06, + "loss": 0.6435, + "step": 1066 + }, + { + "epoch": 0.2536249108628476, + "grad_norm": 1.09016304024847, + "learning_rate": 8.746478831290648e-06, + "loss": 0.6975, + "step": 1067 + }, + { + "epoch": 0.25386260993582127, + "grad_norm": 1.2142712551063848, + "learning_rate": 8.74392812019938e-06, + "loss": 0.6685, + "step": 1068 + }, + { + "epoch": 0.2541003090087949, + "grad_norm": 1.156922396051977, + "learning_rate": 8.741375189343625e-06, + "loss": 0.6136, + "step": 1069 + }, + { + "epoch": 0.2543380080817685, + "grad_norm": 1.141038585979363, + "learning_rate": 8.738820040237002e-06, + "loss": 0.6367, + "step": 1070 + }, + { + "epoch": 0.2545757071547421, + "grad_norm": 1.1452811270404202, + "learning_rate": 8.736262674394455e-06, + "loss": 0.6291, + "step": 1071 + }, + { + "epoch": 0.2548134062277157, + "grad_norm": 1.0981937562531785, + "learning_rate": 8.733703093332237e-06, + "loss": 0.6696, + "step": 1072 + }, + { + "epoch": 0.2550511053006893, + "grad_norm": 1.3220746097011782, + "learning_rate": 8.73114129856792e-06, + "loss": 0.6277, + "step": 1073 + }, + { + "epoch": 0.25528880437366297, + "grad_norm": 1.1914573625218026, + "learning_rate": 8.728577291620381e-06, + "loss": 0.5866, + "step": 1074 + }, + { + "epoch": 0.2555265034466366, + "grad_norm": 1.1265589936073233, + "learning_rate": 8.726011074009813e-06, + "loss": 0.6317, + "step": 1075 + }, + { + "epoch": 0.2557642025196102, + "grad_norm": 1.3745441007762373, + "learning_rate": 8.723442647257723e-06, + "loss": 0.6513, + "step": 1076 + }, + { + "epoch": 0.2560019015925838, + "grad_norm": 1.3387304982947705, + "learning_rate": 8.720872012886918e-06, + "loss": 0.6447, + "step": 1077 + }, + { + "epoch": 0.2562396006655574, + "grad_norm": 1.4068172384202604, + "learning_rate": 8.718299172421525e-06, + "loss": 0.6469, + "step": 1078 + }, + { + "epoch": 0.256477299738531, + "grad_norm": 1.2800298435494148, + "learning_rate": 8.715724127386971e-06, + "loss": 0.6498, + "step": 1079 + }, + { + "epoch": 0.2567149988115046, + "grad_norm": 1.2087370936970054, + "learning_rate": 8.713146879309995e-06, + "loss": 0.6605, + "step": 1080 + }, + { + "epoch": 0.2569526978844783, + "grad_norm": 1.210375005052214, + "learning_rate": 8.71056742971864e-06, + "loss": 0.6411, + "step": 1081 + }, + { + "epoch": 0.2571903969574519, + "grad_norm": 1.3672669166405453, + "learning_rate": 8.707985780142252e-06, + "loss": 0.6171, + "step": 1082 + }, + { + "epoch": 0.2574280960304255, + "grad_norm": 1.5188586721045785, + "learning_rate": 8.705401932111487e-06, + "loss": 0.6313, + "step": 1083 + }, + { + "epoch": 0.2576657951033991, + "grad_norm": 1.9539016637294688, + "learning_rate": 8.702815887158297e-06, + "loss": 0.6535, + "step": 1084 + }, + { + "epoch": 0.2579034941763727, + "grad_norm": 1.2339224023355986, + "learning_rate": 8.700227646815947e-06, + "loss": 0.6262, + "step": 1085 + }, + { + "epoch": 0.2581411932493463, + "grad_norm": 1.5084534028997227, + "learning_rate": 8.697637212618992e-06, + "loss": 0.6159, + "step": 1086 + }, + { + "epoch": 0.2583788923223199, + "grad_norm": 1.2476676096446777, + "learning_rate": 8.695044586103297e-06, + "loss": 0.6306, + "step": 1087 + }, + { + "epoch": 0.2586165913952936, + "grad_norm": 1.3086712725237983, + "learning_rate": 8.69244976880602e-06, + "loss": 0.5796, + "step": 1088 + }, + { + "epoch": 0.2588542904682672, + "grad_norm": 1.210948769822897, + "learning_rate": 8.68985276226562e-06, + "loss": 0.6401, + "step": 1089 + }, + { + "epoch": 0.2590919895412408, + "grad_norm": 1.2911406370006486, + "learning_rate": 8.68725356802186e-06, + "loss": 0.6203, + "step": 1090 + }, + { + "epoch": 0.2593296886142144, + "grad_norm": 1.4579458910453298, + "learning_rate": 8.68465218761579e-06, + "loss": 0.6578, + "step": 1091 + }, + { + "epoch": 0.259567387687188, + "grad_norm": 1.468226500257744, + "learning_rate": 8.682048622589764e-06, + "loss": 0.6025, + "step": 1092 + }, + { + "epoch": 0.2598050867601616, + "grad_norm": 1.927314725293736, + "learning_rate": 8.679442874487427e-06, + "loss": 0.6577, + "step": 1093 + }, + { + "epoch": 0.2600427858331352, + "grad_norm": 1.4417035403628373, + "learning_rate": 8.676834944853722e-06, + "loss": 0.6759, + "step": 1094 + }, + { + "epoch": 0.2602804849061089, + "grad_norm": 1.500793166524623, + "learning_rate": 8.674224835234879e-06, + "loss": 0.618, + "step": 1095 + }, + { + "epoch": 0.2605181839790825, + "grad_norm": 1.264429486051268, + "learning_rate": 8.671612547178428e-06, + "loss": 0.6679, + "step": 1096 + }, + { + "epoch": 0.2607558830520561, + "grad_norm": 1.1934426412072952, + "learning_rate": 8.668998082233186e-06, + "loss": 0.6396, + "step": 1097 + }, + { + "epoch": 0.2609935821250297, + "grad_norm": 2.0553310152766757, + "learning_rate": 8.666381441949263e-06, + "loss": 0.6241, + "step": 1098 + }, + { + "epoch": 0.2612312811980033, + "grad_norm": 1.3781165974132294, + "learning_rate": 8.663762627878059e-06, + "loss": 0.6107, + "step": 1099 + }, + { + "epoch": 0.2614689802709769, + "grad_norm": 1.0783247823958095, + "learning_rate": 8.661141641572256e-06, + "loss": 0.6383, + "step": 1100 + }, + { + "epoch": 0.2614689802709769, + "eval_loss": 0.6430635452270508, + "eval_runtime": 278.4487, + "eval_samples_per_second": 39.066, + "eval_steps_per_second": 1.221, + "step": 1100 + }, + { + "epoch": 0.26170667934395053, + "grad_norm": 1.3356087791546152, + "learning_rate": 8.658518484585834e-06, + "loss": 0.6376, + "step": 1101 + }, + { + "epoch": 0.2619443784169242, + "grad_norm": 1.319340085275578, + "learning_rate": 8.655893158474056e-06, + "loss": 0.6319, + "step": 1102 + }, + { + "epoch": 0.2621820774898978, + "grad_norm": 1.2310005873749101, + "learning_rate": 8.653265664793466e-06, + "loss": 0.6339, + "step": 1103 + }, + { + "epoch": 0.2624197765628714, + "grad_norm": 1.4576520664456325, + "learning_rate": 8.650636005101903e-06, + "loss": 0.6886, + "step": 1104 + }, + { + "epoch": 0.262657475635845, + "grad_norm": 1.552344684348575, + "learning_rate": 8.64800418095848e-06, + "loss": 0.6778, + "step": 1105 + }, + { + "epoch": 0.2628951747088186, + "grad_norm": 1.440005217989413, + "learning_rate": 8.645370193923604e-06, + "loss": 0.6713, + "step": 1106 + }, + { + "epoch": 0.26313287378179223, + "grad_norm": 1.259875690270759, + "learning_rate": 8.642734045558952e-06, + "loss": 0.6363, + "step": 1107 + }, + { + "epoch": 0.26337057285476584, + "grad_norm": 1.1539004728031181, + "learning_rate": 8.640095737427494e-06, + "loss": 0.5957, + "step": 1108 + }, + { + "epoch": 0.2636082719277395, + "grad_norm": 1.3623947752424979, + "learning_rate": 8.637455271093473e-06, + "loss": 0.6704, + "step": 1109 + }, + { + "epoch": 0.2638459710007131, + "grad_norm": 1.203935096968745, + "learning_rate": 8.634812648122413e-06, + "loss": 0.6276, + "step": 1110 + }, + { + "epoch": 0.2640836700736867, + "grad_norm": 1.6766184158038697, + "learning_rate": 8.632167870081122e-06, + "loss": 0.6257, + "step": 1111 + }, + { + "epoch": 0.2643213691466603, + "grad_norm": 1.2620498365753718, + "learning_rate": 8.629520938537677e-06, + "loss": 0.7135, + "step": 1112 + }, + { + "epoch": 0.26455906821963393, + "grad_norm": 1.2319360921403815, + "learning_rate": 8.626871855061438e-06, + "loss": 0.6184, + "step": 1113 + }, + { + "epoch": 0.26479676729260754, + "grad_norm": 1.2502835054265478, + "learning_rate": 8.624220621223041e-06, + "loss": 0.6392, + "step": 1114 + }, + { + "epoch": 0.2650344663655812, + "grad_norm": 1.3120253038123306, + "learning_rate": 8.621567238594392e-06, + "loss": 0.5815, + "step": 1115 + }, + { + "epoch": 0.2652721654385548, + "grad_norm": 1.1706266616761272, + "learning_rate": 8.618911708748677e-06, + "loss": 0.6469, + "step": 1116 + }, + { + "epoch": 0.2655098645115284, + "grad_norm": 1.1521378191507614, + "learning_rate": 8.616254033260351e-06, + "loss": 0.622, + "step": 1117 + }, + { + "epoch": 0.265747563584502, + "grad_norm": 1.9252091969932197, + "learning_rate": 8.61359421370514e-06, + "loss": 0.6534, + "step": 1118 + }, + { + "epoch": 0.26598526265747563, + "grad_norm": 1.2218063275834092, + "learning_rate": 8.610932251660046e-06, + "loss": 0.6204, + "step": 1119 + }, + { + "epoch": 0.26622296173044924, + "grad_norm": 1.1857374576904602, + "learning_rate": 8.60826814870334e-06, + "loss": 0.6404, + "step": 1120 + }, + { + "epoch": 0.26646066080342284, + "grad_norm": 1.1562056790436879, + "learning_rate": 8.60560190641456e-06, + "loss": 0.6209, + "step": 1121 + }, + { + "epoch": 0.2666983598763965, + "grad_norm": 1.3122300341507134, + "learning_rate": 8.602933526374512e-06, + "loss": 0.6253, + "step": 1122 + }, + { + "epoch": 0.2669360589493701, + "grad_norm": 1.1666364441011476, + "learning_rate": 8.600263010165275e-06, + "loss": 0.6853, + "step": 1123 + }, + { + "epoch": 0.2671737580223437, + "grad_norm": 1.1903919747671061, + "learning_rate": 8.597590359370186e-06, + "loss": 0.6228, + "step": 1124 + }, + { + "epoch": 0.26741145709531733, + "grad_norm": 1.2063903052486553, + "learning_rate": 8.594915575573852e-06, + "loss": 0.6394, + "step": 1125 + }, + { + "epoch": 0.26764915616829094, + "grad_norm": 1.480149180008114, + "learning_rate": 8.592238660362149e-06, + "loss": 0.6312, + "step": 1126 + }, + { + "epoch": 0.26788685524126454, + "grad_norm": 1.244172379837697, + "learning_rate": 8.58955961532221e-06, + "loss": 0.6432, + "step": 1127 + }, + { + "epoch": 0.26812455431423815, + "grad_norm": 1.250133915812739, + "learning_rate": 8.586878442042435e-06, + "loss": 0.6149, + "step": 1128 + }, + { + "epoch": 0.2683622533872118, + "grad_norm": 1.1951234596863411, + "learning_rate": 8.584195142112482e-06, + "loss": 0.6239, + "step": 1129 + }, + { + "epoch": 0.2685999524601854, + "grad_norm": 1.21714213030725, + "learning_rate": 8.581509717123272e-06, + "loss": 0.6437, + "step": 1130 + }, + { + "epoch": 0.268837651533159, + "grad_norm": 2.1769166688711694, + "learning_rate": 8.578822168666987e-06, + "loss": 0.6463, + "step": 1131 + }, + { + "epoch": 0.26907535060613264, + "grad_norm": 1.1702813550254079, + "learning_rate": 8.576132498337069e-06, + "loss": 0.6308, + "step": 1132 + }, + { + "epoch": 0.26931304967910624, + "grad_norm": 1.4606715889574786, + "learning_rate": 8.57344070772821e-06, + "loss": 0.6413, + "step": 1133 + }, + { + "epoch": 0.26955074875207985, + "grad_norm": 1.2725774719645704, + "learning_rate": 8.570746798436372e-06, + "loss": 0.652, + "step": 1134 + }, + { + "epoch": 0.26978844782505346, + "grad_norm": 1.185111271357713, + "learning_rate": 8.568050772058763e-06, + "loss": 0.654, + "step": 1135 + }, + { + "epoch": 0.2700261468980271, + "grad_norm": 1.6235069006292686, + "learning_rate": 8.565352630193849e-06, + "loss": 0.6632, + "step": 1136 + }, + { + "epoch": 0.2702638459710007, + "grad_norm": 1.4478916889763318, + "learning_rate": 8.56265237444135e-06, + "loss": 0.6161, + "step": 1137 + }, + { + "epoch": 0.27050154504397433, + "grad_norm": 1.2047007040158266, + "learning_rate": 8.559950006402242e-06, + "loss": 0.6689, + "step": 1138 + }, + { + "epoch": 0.27073924411694794, + "grad_norm": 1.2994566517145638, + "learning_rate": 8.557245527678753e-06, + "loss": 0.6272, + "step": 1139 + }, + { + "epoch": 0.27097694318992155, + "grad_norm": 1.7809036959064193, + "learning_rate": 8.554538939874356e-06, + "loss": 0.6323, + "step": 1140 + }, + { + "epoch": 0.27121464226289516, + "grad_norm": 1.4390114099280487, + "learning_rate": 8.551830244593785e-06, + "loss": 0.6391, + "step": 1141 + }, + { + "epoch": 0.27145234133586876, + "grad_norm": 1.3642353750287868, + "learning_rate": 8.549119443443013e-06, + "loss": 0.6293, + "step": 1142 + }, + { + "epoch": 0.2716900404088424, + "grad_norm": 1.2662308245696157, + "learning_rate": 8.546406538029268e-06, + "loss": 0.6672, + "step": 1143 + }, + { + "epoch": 0.27192773948181603, + "grad_norm": 1.2902669446203519, + "learning_rate": 8.543691529961025e-06, + "loss": 0.5805, + "step": 1144 + }, + { + "epoch": 0.27216543855478964, + "grad_norm": 1.184885104687112, + "learning_rate": 8.540974420848004e-06, + "loss": 0.6164, + "step": 1145 + }, + { + "epoch": 0.27240313762776325, + "grad_norm": 1.2203628759342886, + "learning_rate": 8.538255212301173e-06, + "loss": 0.6763, + "step": 1146 + }, + { + "epoch": 0.27264083670073685, + "grad_norm": 1.2946290261001958, + "learning_rate": 8.535533905932739e-06, + "loss": 0.654, + "step": 1147 + }, + { + "epoch": 0.27287853577371046, + "grad_norm": 1.1048285462465834, + "learning_rate": 8.532810503356159e-06, + "loss": 0.5973, + "step": 1148 + }, + { + "epoch": 0.2731162348466841, + "grad_norm": 1.4163821332899769, + "learning_rate": 8.530085006186133e-06, + "loss": 0.7318, + "step": 1149 + }, + { + "epoch": 0.27335393391965773, + "grad_norm": 1.5523182921408571, + "learning_rate": 8.527357416038599e-06, + "loss": 0.613, + "step": 1150 + }, + { + "epoch": 0.27359163299263134, + "grad_norm": 1.4135485502859797, + "learning_rate": 8.524627734530738e-06, + "loss": 0.6407, + "step": 1151 + }, + { + "epoch": 0.27382933206560495, + "grad_norm": 1.197462754397783, + "learning_rate": 8.521895963280967e-06, + "loss": 0.6305, + "step": 1152 + }, + { + "epoch": 0.27406703113857855, + "grad_norm": 1.300459769572292, + "learning_rate": 8.519162103908951e-06, + "loss": 0.6081, + "step": 1153 + }, + { + "epoch": 0.27430473021155216, + "grad_norm": 1.3539675232544177, + "learning_rate": 8.516426158035584e-06, + "loss": 0.657, + "step": 1154 + }, + { + "epoch": 0.27454242928452577, + "grad_norm": 1.3596956592450349, + "learning_rate": 8.513688127283002e-06, + "loss": 0.6249, + "step": 1155 + }, + { + "epoch": 0.27478012835749943, + "grad_norm": 1.335892520485724, + "learning_rate": 8.510948013274575e-06, + "loss": 0.6314, + "step": 1156 + }, + { + "epoch": 0.27501782743047304, + "grad_norm": 1.312728277612895, + "learning_rate": 8.508205817634908e-06, + "loss": 0.6837, + "step": 1157 + }, + { + "epoch": 0.27525552650344665, + "grad_norm": 1.248437694574924, + "learning_rate": 8.505461541989846e-06, + "loss": 0.6092, + "step": 1158 + }, + { + "epoch": 0.27549322557642025, + "grad_norm": 1.2011868056107662, + "learning_rate": 8.502715187966455e-06, + "loss": 0.6476, + "step": 1159 + }, + { + "epoch": 0.27573092464939386, + "grad_norm": 1.1885821011908153, + "learning_rate": 8.499966757193047e-06, + "loss": 0.6614, + "step": 1160 + }, + { + "epoch": 0.27596862372236747, + "grad_norm": 1.1583963478283392, + "learning_rate": 8.497216251299155e-06, + "loss": 0.6517, + "step": 1161 + }, + { + "epoch": 0.2762063227953411, + "grad_norm": 1.5115279078574175, + "learning_rate": 8.494463671915547e-06, + "loss": 0.6035, + "step": 1162 + }, + { + "epoch": 0.27644402186831474, + "grad_norm": 1.2881505191138805, + "learning_rate": 8.491709020674221e-06, + "loss": 0.576, + "step": 1163 + }, + { + "epoch": 0.27668172094128834, + "grad_norm": 2.7132625735465354, + "learning_rate": 8.488952299208402e-06, + "loss": 0.6726, + "step": 1164 + }, + { + "epoch": 0.27691942001426195, + "grad_norm": 1.1547488351879804, + "learning_rate": 8.48619350915254e-06, + "loss": 0.5937, + "step": 1165 + }, + { + "epoch": 0.27715711908723556, + "grad_norm": 1.3540492999543747, + "learning_rate": 8.483432652142316e-06, + "loss": 0.6401, + "step": 1166 + }, + { + "epoch": 0.27739481816020917, + "grad_norm": 1.2524436158025514, + "learning_rate": 8.480669729814635e-06, + "loss": 0.6731, + "step": 1167 + }, + { + "epoch": 0.2776325172331828, + "grad_norm": 1.2400752277923066, + "learning_rate": 8.477904743807622e-06, + "loss": 0.6358, + "step": 1168 + }, + { + "epoch": 0.2778702163061564, + "grad_norm": 1.4844628410828986, + "learning_rate": 8.475137695760634e-06, + "loss": 0.6217, + "step": 1169 + }, + { + "epoch": 0.27810791537913004, + "grad_norm": 1.3591211981978755, + "learning_rate": 8.472368587314244e-06, + "loss": 0.6511, + "step": 1170 + }, + { + "epoch": 0.27834561445210365, + "grad_norm": 1.3128555443473522, + "learning_rate": 8.469597420110249e-06, + "loss": 0.6878, + "step": 1171 + }, + { + "epoch": 0.27858331352507726, + "grad_norm": 1.2308976957366993, + "learning_rate": 8.466824195791665e-06, + "loss": 0.6374, + "step": 1172 + }, + { + "epoch": 0.27882101259805087, + "grad_norm": 1.2666198616061923, + "learning_rate": 8.464048916002732e-06, + "loss": 0.6733, + "step": 1173 + }, + { + "epoch": 0.2790587116710245, + "grad_norm": 1.1941214765817123, + "learning_rate": 8.461271582388903e-06, + "loss": 0.6377, + "step": 1174 + }, + { + "epoch": 0.2792964107439981, + "grad_norm": 1.2940512125694017, + "learning_rate": 8.458492196596852e-06, + "loss": 0.6478, + "step": 1175 + }, + { + "epoch": 0.2795341098169717, + "grad_norm": 1.164183896649691, + "learning_rate": 8.455710760274468e-06, + "loss": 0.6349, + "step": 1176 + }, + { + "epoch": 0.27977180888994535, + "grad_norm": 1.227867015484948, + "learning_rate": 8.452927275070858e-06, + "loss": 0.6384, + "step": 1177 + }, + { + "epoch": 0.28000950796291896, + "grad_norm": 1.124879322043426, + "learning_rate": 8.450141742636343e-06, + "loss": 0.6054, + "step": 1178 + }, + { + "epoch": 0.28024720703589256, + "grad_norm": 1.1634457325903744, + "learning_rate": 8.447354164622457e-06, + "loss": 0.6851, + "step": 1179 + }, + { + "epoch": 0.28048490610886617, + "grad_norm": 1.209552814284204, + "learning_rate": 8.444564542681946e-06, + "loss": 0.6149, + "step": 1180 + }, + { + "epoch": 0.2807226051818398, + "grad_norm": 1.1869421357048744, + "learning_rate": 8.44177287846877e-06, + "loss": 0.6253, + "step": 1181 + }, + { + "epoch": 0.2809603042548134, + "grad_norm": 1.2626398362217186, + "learning_rate": 8.4389791736381e-06, + "loss": 0.6124, + "step": 1182 + }, + { + "epoch": 0.281198003327787, + "grad_norm": 1.2537261849126289, + "learning_rate": 8.436183429846314e-06, + "loss": 0.645, + "step": 1183 + }, + { + "epoch": 0.28143570240076066, + "grad_norm": 1.138966277901728, + "learning_rate": 8.433385648751003e-06, + "loss": 0.5974, + "step": 1184 + }, + { + "epoch": 0.28167340147373426, + "grad_norm": 1.306726825087157, + "learning_rate": 8.43058583201096e-06, + "loss": 0.6266, + "step": 1185 + }, + { + "epoch": 0.28191110054670787, + "grad_norm": 1.1402207024575646, + "learning_rate": 8.42778398128619e-06, + "loss": 0.6352, + "step": 1186 + }, + { + "epoch": 0.2821487996196815, + "grad_norm": 1.5591428467431343, + "learning_rate": 8.424980098237904e-06, + "loss": 0.7008, + "step": 1187 + }, + { + "epoch": 0.2823864986926551, + "grad_norm": 1.1343557316630457, + "learning_rate": 8.422174184528514e-06, + "loss": 0.5937, + "step": 1188 + }, + { + "epoch": 0.2826241977656287, + "grad_norm": 1.2236860876725582, + "learning_rate": 8.41936624182164e-06, + "loss": 0.6447, + "step": 1189 + }, + { + "epoch": 0.28286189683860236, + "grad_norm": 1.1650475622111356, + "learning_rate": 8.416556271782103e-06, + "loss": 0.6125, + "step": 1190 + }, + { + "epoch": 0.28309959591157596, + "grad_norm": 1.2196445069340456, + "learning_rate": 8.413744276075928e-06, + "loss": 0.6205, + "step": 1191 + }, + { + "epoch": 0.28333729498454957, + "grad_norm": 1.2487977415460816, + "learning_rate": 8.410930256370337e-06, + "loss": 0.5924, + "step": 1192 + }, + { + "epoch": 0.2835749940575232, + "grad_norm": 1.2653081652653535, + "learning_rate": 8.408114214333754e-06, + "loss": 0.6208, + "step": 1193 + }, + { + "epoch": 0.2838126931304968, + "grad_norm": 1.3164138531158058, + "learning_rate": 8.405296151635804e-06, + "loss": 0.6636, + "step": 1194 + }, + { + "epoch": 0.2840503922034704, + "grad_norm": 1.256816531580835, + "learning_rate": 8.402476069947309e-06, + "loss": 0.6616, + "step": 1195 + }, + { + "epoch": 0.284288091276444, + "grad_norm": 1.2119437727090459, + "learning_rate": 8.399653970940284e-06, + "loss": 0.6197, + "step": 1196 + }, + { + "epoch": 0.28452579034941766, + "grad_norm": 1.202603586927377, + "learning_rate": 8.39682985628795e-06, + "loss": 0.6285, + "step": 1197 + }, + { + "epoch": 0.28476348942239127, + "grad_norm": 1.2663403697234838, + "learning_rate": 8.39400372766471e-06, + "loss": 0.6852, + "step": 1198 + }, + { + "epoch": 0.2850011884953649, + "grad_norm": 1.2612976390318162, + "learning_rate": 8.39117558674617e-06, + "loss": 0.6343, + "step": 1199 + }, + { + "epoch": 0.2852388875683385, + "grad_norm": 1.1596780582414723, + "learning_rate": 8.388345435209129e-06, + "loss": 0.6338, + "step": 1200 + }, + { + "epoch": 0.2852388875683385, + "eval_loss": 0.6375774145126343, + "eval_runtime": 278.7804, + "eval_samples_per_second": 39.02, + "eval_steps_per_second": 1.22, + "step": 1200 + }, + { + "epoch": 0.2854765866413121, + "grad_norm": 1.297600893579044, + "learning_rate": 8.385513274731574e-06, + "loss": 0.6069, + "step": 1201 + }, + { + "epoch": 0.2857142857142857, + "grad_norm": 1.246827055093378, + "learning_rate": 8.382679106992687e-06, + "loss": 0.5648, + "step": 1202 + }, + { + "epoch": 0.2859519847872593, + "grad_norm": 1.1607212156928877, + "learning_rate": 8.379842933672833e-06, + "loss": 0.6362, + "step": 1203 + }, + { + "epoch": 0.28618968386023297, + "grad_norm": 1.4035765770864534, + "learning_rate": 8.377004756453575e-06, + "loss": 0.6354, + "step": 1204 + }, + { + "epoch": 0.2864273829332066, + "grad_norm": 1.3932872655198736, + "learning_rate": 8.374164577017663e-06, + "loss": 0.6278, + "step": 1205 + }, + { + "epoch": 0.2866650820061802, + "grad_norm": 2.259180145009223, + "learning_rate": 8.371322397049026e-06, + "loss": 0.6078, + "step": 1206 + }, + { + "epoch": 0.2869027810791538, + "grad_norm": 1.3345229098453393, + "learning_rate": 8.368478218232787e-06, + "loss": 0.6363, + "step": 1207 + }, + { + "epoch": 0.2871404801521274, + "grad_norm": 1.2692360708095929, + "learning_rate": 8.365632042255253e-06, + "loss": 0.6232, + "step": 1208 + }, + { + "epoch": 0.287378179225101, + "grad_norm": 1.119930478216007, + "learning_rate": 8.36278387080391e-06, + "loss": 0.6029, + "step": 1209 + }, + { + "epoch": 0.2876158782980746, + "grad_norm": 1.1351268391361722, + "learning_rate": 8.359933705567435e-06, + "loss": 0.6545, + "step": 1210 + }, + { + "epoch": 0.2878535773710483, + "grad_norm": 1.1977219525334828, + "learning_rate": 8.35708154823568e-06, + "loss": 0.6186, + "step": 1211 + }, + { + "epoch": 0.2880912764440219, + "grad_norm": 1.771679939983377, + "learning_rate": 8.354227400499684e-06, + "loss": 0.6913, + "step": 1212 + }, + { + "epoch": 0.2883289755169955, + "grad_norm": 1.2110106973632386, + "learning_rate": 8.351371264051659e-06, + "loss": 0.6266, + "step": 1213 + }, + { + "epoch": 0.2885666745899691, + "grad_norm": 1.1725653306548807, + "learning_rate": 8.348513140585005e-06, + "loss": 0.6495, + "step": 1214 + }, + { + "epoch": 0.2888043736629427, + "grad_norm": 1.1662214640439534, + "learning_rate": 8.345653031794292e-06, + "loss": 0.6757, + "step": 1215 + }, + { + "epoch": 0.2890420727359163, + "grad_norm": 1.3032368637587195, + "learning_rate": 8.342790939375271e-06, + "loss": 0.6434, + "step": 1216 + }, + { + "epoch": 0.2892797718088899, + "grad_norm": 1.0825177779065902, + "learning_rate": 8.339926865024871e-06, + "loss": 0.6316, + "step": 1217 + }, + { + "epoch": 0.2895174708818636, + "grad_norm": 1.24211567501475, + "learning_rate": 8.33706081044119e-06, + "loss": 0.5944, + "step": 1218 + }, + { + "epoch": 0.2897551699548372, + "grad_norm": 1.1401800657619263, + "learning_rate": 8.334192777323508e-06, + "loss": 0.6403, + "step": 1219 + }, + { + "epoch": 0.2899928690278108, + "grad_norm": 1.2897211236924202, + "learning_rate": 8.33132276737227e-06, + "loss": 0.6372, + "step": 1220 + }, + { + "epoch": 0.2902305681007844, + "grad_norm": 1.3732479561584627, + "learning_rate": 8.328450782289098e-06, + "loss": 0.5951, + "step": 1221 + }, + { + "epoch": 0.290468267173758, + "grad_norm": 1.7032877529908732, + "learning_rate": 8.325576823776785e-06, + "loss": 0.6579, + "step": 1222 + }, + { + "epoch": 0.2907059662467316, + "grad_norm": 1.1522717321442797, + "learning_rate": 8.32270089353929e-06, + "loss": 0.6067, + "step": 1223 + }, + { + "epoch": 0.2909436653197053, + "grad_norm": 1.3478238900476383, + "learning_rate": 8.319822993281749e-06, + "loss": 0.5874, + "step": 1224 + }, + { + "epoch": 0.2911813643926789, + "grad_norm": 1.5425964613120116, + "learning_rate": 8.316943124710457e-06, + "loss": 0.6094, + "step": 1225 + }, + { + "epoch": 0.2914190634656525, + "grad_norm": 1.3072095347359909, + "learning_rate": 8.31406128953288e-06, + "loss": 0.6554, + "step": 1226 + }, + { + "epoch": 0.2916567625386261, + "grad_norm": 1.3214082655350412, + "learning_rate": 8.311177489457653e-06, + "loss": 0.6543, + "step": 1227 + }, + { + "epoch": 0.2918944616115997, + "grad_norm": 1.2207665370204042, + "learning_rate": 8.30829172619457e-06, + "loss": 0.6568, + "step": 1228 + }, + { + "epoch": 0.2921321606845733, + "grad_norm": 1.2773380904588207, + "learning_rate": 8.305404001454592e-06, + "loss": 0.6643, + "step": 1229 + }, + { + "epoch": 0.2923698597575469, + "grad_norm": 1.1640307634838711, + "learning_rate": 8.302514316949846e-06, + "loss": 0.6241, + "step": 1230 + }, + { + "epoch": 0.2926075588305206, + "grad_norm": 1.5506455258213687, + "learning_rate": 8.299622674393615e-06, + "loss": 0.6024, + "step": 1231 + }, + { + "epoch": 0.2928452579034942, + "grad_norm": 1.1717557545980903, + "learning_rate": 8.296729075500345e-06, + "loss": 0.59, + "step": 1232 + }, + { + "epoch": 0.2930829569764678, + "grad_norm": 1.2193042090083075, + "learning_rate": 8.293833521985647e-06, + "loss": 0.6122, + "step": 1233 + }, + { + "epoch": 0.2933206560494414, + "grad_norm": 1.1592580479889718, + "learning_rate": 8.290936015566283e-06, + "loss": 0.5882, + "step": 1234 + }, + { + "epoch": 0.293558355122415, + "grad_norm": 1.230241034526615, + "learning_rate": 8.288036557960175e-06, + "loss": 0.632, + "step": 1235 + }, + { + "epoch": 0.2937960541953886, + "grad_norm": 1.3854899211363627, + "learning_rate": 8.285135150886408e-06, + "loss": 0.5784, + "step": 1236 + }, + { + "epoch": 0.29403375326836223, + "grad_norm": 1.2940116928148993, + "learning_rate": 8.282231796065215e-06, + "loss": 0.6283, + "step": 1237 + }, + { + "epoch": 0.2942714523413359, + "grad_norm": 1.2824891482851641, + "learning_rate": 8.279326495217987e-06, + "loss": 0.6, + "step": 1238 + }, + { + "epoch": 0.2945091514143095, + "grad_norm": 1.183207454002302, + "learning_rate": 8.27641925006727e-06, + "loss": 0.5953, + "step": 1239 + }, + { + "epoch": 0.2947468504872831, + "grad_norm": 2.0638795668833163, + "learning_rate": 8.27351006233676e-06, + "loss": 0.6893, + "step": 1240 + }, + { + "epoch": 0.2949845495602567, + "grad_norm": 1.2080979284058333, + "learning_rate": 8.270598933751308e-06, + "loss": 0.6423, + "step": 1241 + }, + { + "epoch": 0.2952222486332303, + "grad_norm": 1.223581884279364, + "learning_rate": 8.267685866036912e-06, + "loss": 0.6519, + "step": 1242 + }, + { + "epoch": 0.29545994770620393, + "grad_norm": 1.2364284698190002, + "learning_rate": 8.264770860920722e-06, + "loss": 0.6433, + "step": 1243 + }, + { + "epoch": 0.29569764677917754, + "grad_norm": 1.0895984249031034, + "learning_rate": 8.26185392013104e-06, + "loss": 0.662, + "step": 1244 + }, + { + "epoch": 0.2959353458521512, + "grad_norm": 1.2242620757974114, + "learning_rate": 8.258935045397306e-06, + "loss": 0.6066, + "step": 1245 + }, + { + "epoch": 0.2961730449251248, + "grad_norm": 1.2291262265816612, + "learning_rate": 8.256014238450117e-06, + "loss": 0.6575, + "step": 1246 + }, + { + "epoch": 0.2964107439980984, + "grad_norm": 1.2840889677441565, + "learning_rate": 8.25309150102121e-06, + "loss": 0.6626, + "step": 1247 + }, + { + "epoch": 0.296648443071072, + "grad_norm": 1.2405884379047993, + "learning_rate": 8.250166834843469e-06, + "loss": 0.6147, + "step": 1248 + }, + { + "epoch": 0.29688614214404563, + "grad_norm": 1.4488646232307982, + "learning_rate": 8.247240241650918e-06, + "loss": 0.6217, + "step": 1249 + }, + { + "epoch": 0.29712384121701924, + "grad_norm": 1.235418540365753, + "learning_rate": 8.244311723178728e-06, + "loss": 0.6481, + "step": 1250 + }, + { + "epoch": 0.29736154028999284, + "grad_norm": 1.327562375277948, + "learning_rate": 8.241381281163207e-06, + "loss": 0.5943, + "step": 1251 + }, + { + "epoch": 0.2975992393629665, + "grad_norm": 1.2133606869308535, + "learning_rate": 8.23844891734181e-06, + "loss": 0.6248, + "step": 1252 + }, + { + "epoch": 0.2978369384359401, + "grad_norm": 2.484132033866223, + "learning_rate": 8.235514633453123e-06, + "loss": 0.5903, + "step": 1253 + }, + { + "epoch": 0.2980746375089137, + "grad_norm": 1.2046983965154663, + "learning_rate": 8.232578431236877e-06, + "loss": 0.626, + "step": 1254 + }, + { + "epoch": 0.2983123365818873, + "grad_norm": 1.3269126535718765, + "learning_rate": 8.229640312433938e-06, + "loss": 0.5699, + "step": 1255 + }, + { + "epoch": 0.29855003565486093, + "grad_norm": 1.214882557131175, + "learning_rate": 8.226700278786306e-06, + "loss": 0.6099, + "step": 1256 + }, + { + "epoch": 0.29878773472783454, + "grad_norm": 1.2098288818709817, + "learning_rate": 8.223758332037121e-06, + "loss": 0.6315, + "step": 1257 + }, + { + "epoch": 0.29902543380080815, + "grad_norm": 1.2338930625206423, + "learning_rate": 8.220814473930656e-06, + "loss": 0.5846, + "step": 1258 + }, + { + "epoch": 0.2992631328737818, + "grad_norm": 1.660283067788018, + "learning_rate": 8.217868706212313e-06, + "loss": 0.6196, + "step": 1259 + }, + { + "epoch": 0.2995008319467554, + "grad_norm": 1.7316171574248282, + "learning_rate": 8.214921030628632e-06, + "loss": 0.6532, + "step": 1260 + }, + { + "epoch": 0.299738531019729, + "grad_norm": 1.3159461919212463, + "learning_rate": 8.21197144892728e-06, + "loss": 0.6022, + "step": 1261 + }, + { + "epoch": 0.29997623009270263, + "grad_norm": 1.2005698047243003, + "learning_rate": 8.209019962857058e-06, + "loss": 0.5741, + "step": 1262 + }, + { + "epoch": 0.30021392916567624, + "grad_norm": 1.2725310990785834, + "learning_rate": 8.206066574167893e-06, + "loss": 0.6084, + "step": 1263 + }, + { + "epoch": 0.30045162823864985, + "grad_norm": 1.1721545484336542, + "learning_rate": 8.203111284610839e-06, + "loss": 0.5875, + "step": 1264 + }, + { + "epoch": 0.3006893273116235, + "grad_norm": 1.1231605326942218, + "learning_rate": 8.20015409593808e-06, + "loss": 0.5921, + "step": 1265 + }, + { + "epoch": 0.3009270263845971, + "grad_norm": 1.1984029379060288, + "learning_rate": 8.197195009902924e-06, + "loss": 0.5992, + "step": 1266 + }, + { + "epoch": 0.3011647254575707, + "grad_norm": 1.1243918630926626, + "learning_rate": 8.194234028259806e-06, + "loss": 0.6704, + "step": 1267 + }, + { + "epoch": 0.30140242453054433, + "grad_norm": 1.8062841159986387, + "learning_rate": 8.191271152764286e-06, + "loss": 0.6154, + "step": 1268 + }, + { + "epoch": 0.30164012360351794, + "grad_norm": 1.3012575099215309, + "learning_rate": 8.18830638517304e-06, + "loss": 0.6322, + "step": 1269 + }, + { + "epoch": 0.30187782267649155, + "grad_norm": 1.2206567914277417, + "learning_rate": 8.185339727243874e-06, + "loss": 0.5748, + "step": 1270 + }, + { + "epoch": 0.30211552174946515, + "grad_norm": 1.3546257191962914, + "learning_rate": 8.182371180735708e-06, + "loss": 0.6363, + "step": 1271 + }, + { + "epoch": 0.3023532208224388, + "grad_norm": 1.39141945647161, + "learning_rate": 8.179400747408586e-06, + "loss": 0.6135, + "step": 1272 + }, + { + "epoch": 0.3025909198954124, + "grad_norm": 2.0192432672079765, + "learning_rate": 8.176428429023674e-06, + "loss": 0.5926, + "step": 1273 + }, + { + "epoch": 0.30282861896838603, + "grad_norm": 1.8343277053751872, + "learning_rate": 8.173454227343243e-06, + "loss": 0.6352, + "step": 1274 + }, + { + "epoch": 0.30306631804135964, + "grad_norm": 1.2643495108112912, + "learning_rate": 8.170478144130695e-06, + "loss": 0.6054, + "step": 1275 + }, + { + "epoch": 0.30330401711433325, + "grad_norm": 1.2684883908004845, + "learning_rate": 8.167500181150541e-06, + "loss": 0.5901, + "step": 1276 + }, + { + "epoch": 0.30354171618730685, + "grad_norm": 1.1040893762738744, + "learning_rate": 8.164520340168404e-06, + "loss": 0.6265, + "step": 1277 + }, + { + "epoch": 0.30377941526028046, + "grad_norm": 1.3000984777253475, + "learning_rate": 8.161538622951023e-06, + "loss": 0.6068, + "step": 1278 + }, + { + "epoch": 0.3040171143332541, + "grad_norm": 1.3627863963801135, + "learning_rate": 8.158555031266255e-06, + "loss": 0.585, + "step": 1279 + }, + { + "epoch": 0.30425481340622773, + "grad_norm": 1.086941239479576, + "learning_rate": 8.15556956688306e-06, + "loss": 0.5616, + "step": 1280 + }, + { + "epoch": 0.30449251247920134, + "grad_norm": 1.4239642935615044, + "learning_rate": 8.15258223157151e-06, + "loss": 0.6388, + "step": 1281 + }, + { + "epoch": 0.30473021155217495, + "grad_norm": 1.2195054786963206, + "learning_rate": 8.14959302710279e-06, + "loss": 0.6164, + "step": 1282 + }, + { + "epoch": 0.30496791062514855, + "grad_norm": 1.3412866044918708, + "learning_rate": 8.146601955249187e-06, + "loss": 0.6083, + "step": 1283 + }, + { + "epoch": 0.30520560969812216, + "grad_norm": 1.460196797877609, + "learning_rate": 8.143609017784106e-06, + "loss": 0.6228, + "step": 1284 + }, + { + "epoch": 0.30544330877109577, + "grad_norm": 1.636977565499973, + "learning_rate": 8.140614216482046e-06, + "loss": 0.6313, + "step": 1285 + }, + { + "epoch": 0.30568100784406943, + "grad_norm": 1.2695608873789355, + "learning_rate": 8.137617553118618e-06, + "loss": 0.5866, + "step": 1286 + }, + { + "epoch": 0.30591870691704304, + "grad_norm": 1.1987584236037503, + "learning_rate": 8.134619029470535e-06, + "loss": 0.6232, + "step": 1287 + }, + { + "epoch": 0.30615640599001664, + "grad_norm": 1.1723083449404061, + "learning_rate": 8.131618647315612e-06, + "loss": 0.6259, + "step": 1288 + }, + { + "epoch": 0.30639410506299025, + "grad_norm": 1.0865768609323623, + "learning_rate": 8.128616408432768e-06, + "loss": 0.6243, + "step": 1289 + }, + { + "epoch": 0.30663180413596386, + "grad_norm": 1.393661223810298, + "learning_rate": 8.125612314602025e-06, + "loss": 0.6866, + "step": 1290 + }, + { + "epoch": 0.30686950320893747, + "grad_norm": 1.2732296594031194, + "learning_rate": 8.122606367604497e-06, + "loss": 0.6053, + "step": 1291 + }, + { + "epoch": 0.3071072022819111, + "grad_norm": 1.3721825502449172, + "learning_rate": 8.119598569222404e-06, + "loss": 0.6143, + "step": 1292 + }, + { + "epoch": 0.30734490135488474, + "grad_norm": 1.1597852058540103, + "learning_rate": 8.116588921239061e-06, + "loss": 0.6347, + "step": 1293 + }, + { + "epoch": 0.30758260042785834, + "grad_norm": 1.3603912177863382, + "learning_rate": 8.113577425438881e-06, + "loss": 0.5765, + "step": 1294 + }, + { + "epoch": 0.30782029950083195, + "grad_norm": 1.3041548116581365, + "learning_rate": 8.110564083607371e-06, + "loss": 0.6202, + "step": 1295 + }, + { + "epoch": 0.30805799857380556, + "grad_norm": 1.201443435627786, + "learning_rate": 8.10754889753113e-06, + "loss": 0.5849, + "step": 1296 + }, + { + "epoch": 0.30829569764677917, + "grad_norm": 1.3934048814850168, + "learning_rate": 8.104531868997858e-06, + "loss": 0.6048, + "step": 1297 + }, + { + "epoch": 0.3085333967197528, + "grad_norm": 1.078565464416661, + "learning_rate": 8.101512999796339e-06, + "loss": 0.5678, + "step": 1298 + }, + { + "epoch": 0.30877109579272644, + "grad_norm": 1.0639279463722897, + "learning_rate": 8.098492291716453e-06, + "loss": 0.598, + "step": 1299 + }, + { + "epoch": 0.30900879486570004, + "grad_norm": 1.135055018907322, + "learning_rate": 8.095469746549172e-06, + "loss": 0.6228, + "step": 1300 + }, + { + "epoch": 0.30900879486570004, + "eval_loss": 0.6324618458747864, + "eval_runtime": 283.8402, + "eval_samples_per_second": 38.324, + "eval_steps_per_second": 1.198, + "step": 1300 + }, + { + "epoch": 0.30924649393867365, + "grad_norm": 1.1621190011407865, + "learning_rate": 8.09244536608655e-06, + "loss": 0.6155, + "step": 1301 + }, + { + "epoch": 0.30948419301164726, + "grad_norm": 1.4699658121598922, + "learning_rate": 8.089419152121736e-06, + "loss": 0.6488, + "step": 1302 + }, + { + "epoch": 0.30972189208462086, + "grad_norm": 1.2090059622378637, + "learning_rate": 8.086391106448965e-06, + "loss": 0.6311, + "step": 1303 + }, + { + "epoch": 0.30995959115759447, + "grad_norm": 1.3386205866440195, + "learning_rate": 8.083361230863557e-06, + "loss": 0.6604, + "step": 1304 + }, + { + "epoch": 0.3101972902305681, + "grad_norm": 1.1090875209677917, + "learning_rate": 8.080329527161914e-06, + "loss": 0.6252, + "step": 1305 + }, + { + "epoch": 0.31043498930354174, + "grad_norm": 1.2962251755065237, + "learning_rate": 8.077295997141525e-06, + "loss": 0.6112, + "step": 1306 + }, + { + "epoch": 0.31067268837651535, + "grad_norm": 1.1953505367884232, + "learning_rate": 8.074260642600963e-06, + "loss": 0.6156, + "step": 1307 + }, + { + "epoch": 0.31091038744948896, + "grad_norm": 1.181709999682525, + "learning_rate": 8.071223465339882e-06, + "loss": 0.6534, + "step": 1308 + }, + { + "epoch": 0.31114808652246256, + "grad_norm": 1.1305420090017293, + "learning_rate": 8.068184467159014e-06, + "loss": 0.6602, + "step": 1309 + }, + { + "epoch": 0.31138578559543617, + "grad_norm": 1.2200347650131202, + "learning_rate": 8.065143649860172e-06, + "loss": 0.6413, + "step": 1310 + }, + { + "epoch": 0.3116234846684098, + "grad_norm": 1.2485671800614122, + "learning_rate": 8.06210101524625e-06, + "loss": 0.6095, + "step": 1311 + }, + { + "epoch": 0.3118611837413834, + "grad_norm": 1.2447228727929402, + "learning_rate": 8.059056565121218e-06, + "loss": 0.6083, + "step": 1312 + }, + { + "epoch": 0.31209888281435705, + "grad_norm": 1.2560964215006982, + "learning_rate": 8.056010301290119e-06, + "loss": 0.5694, + "step": 1313 + }, + { + "epoch": 0.31233658188733066, + "grad_norm": 1.2634767001854996, + "learning_rate": 8.052962225559075e-06, + "loss": 0.5991, + "step": 1314 + }, + { + "epoch": 0.31257428096030426, + "grad_norm": 1.5078761051311538, + "learning_rate": 8.049912339735284e-06, + "loss": 0.6493, + "step": 1315 + }, + { + "epoch": 0.31281198003327787, + "grad_norm": 1.6911150777377995, + "learning_rate": 8.046860645627013e-06, + "loss": 0.6155, + "step": 1316 + }, + { + "epoch": 0.3130496791062515, + "grad_norm": 1.1820611088554882, + "learning_rate": 8.043807145043604e-06, + "loss": 0.5986, + "step": 1317 + }, + { + "epoch": 0.3132873781792251, + "grad_norm": 1.2120048820687497, + "learning_rate": 8.040751839795466e-06, + "loss": 0.6842, + "step": 1318 + }, + { + "epoch": 0.3135250772521987, + "grad_norm": 1.2448271220200928, + "learning_rate": 8.037694731694085e-06, + "loss": 0.6781, + "step": 1319 + }, + { + "epoch": 0.31376277632517235, + "grad_norm": 1.3052592024977614, + "learning_rate": 8.034635822552008e-06, + "loss": 0.6651, + "step": 1320 + }, + { + "epoch": 0.31400047539814596, + "grad_norm": 1.2012539724694387, + "learning_rate": 8.031575114182856e-06, + "loss": 0.6211, + "step": 1321 + }, + { + "epoch": 0.31423817447111957, + "grad_norm": 1.103869303210163, + "learning_rate": 8.028512608401315e-06, + "loss": 0.6197, + "step": 1322 + }, + { + "epoch": 0.3144758735440932, + "grad_norm": 1.3034085080821851, + "learning_rate": 8.025448307023135e-06, + "loss": 0.6524, + "step": 1323 + }, + { + "epoch": 0.3147135726170668, + "grad_norm": 1.2281277240636066, + "learning_rate": 8.022382211865131e-06, + "loss": 0.6286, + "step": 1324 + }, + { + "epoch": 0.3149512716900404, + "grad_norm": 1.310794987112678, + "learning_rate": 8.019314324745183e-06, + "loss": 0.6486, + "step": 1325 + }, + { + "epoch": 0.315188970763014, + "grad_norm": 1.077622531550365, + "learning_rate": 8.016244647482232e-06, + "loss": 0.6522, + "step": 1326 + }, + { + "epoch": 0.31542666983598766, + "grad_norm": 1.2780686092787263, + "learning_rate": 8.013173181896283e-06, + "loss": 0.6341, + "step": 1327 + }, + { + "epoch": 0.31566436890896127, + "grad_norm": 1.2748411716981083, + "learning_rate": 8.010099929808397e-06, + "loss": 0.6885, + "step": 1328 + }, + { + "epoch": 0.3159020679819349, + "grad_norm": 1.5261958237446502, + "learning_rate": 8.007024893040698e-06, + "loss": 0.6514, + "step": 1329 + }, + { + "epoch": 0.3161397670549085, + "grad_norm": 1.193265176939784, + "learning_rate": 8.003948073416365e-06, + "loss": 0.6181, + "step": 1330 + }, + { + "epoch": 0.3163774661278821, + "grad_norm": 1.1823388178692231, + "learning_rate": 8.000869472759639e-06, + "loss": 0.6243, + "step": 1331 + }, + { + "epoch": 0.3166151652008557, + "grad_norm": 1.5577211880406694, + "learning_rate": 7.997789092895809e-06, + "loss": 0.6157, + "step": 1332 + }, + { + "epoch": 0.3168528642738293, + "grad_norm": 1.2460506275616483, + "learning_rate": 7.994706935651228e-06, + "loss": 0.6079, + "step": 1333 + }, + { + "epoch": 0.31709056334680297, + "grad_norm": 1.2648570771980514, + "learning_rate": 7.991623002853296e-06, + "loss": 0.6153, + "step": 1334 + }, + { + "epoch": 0.3173282624197766, + "grad_norm": 1.242960049832451, + "learning_rate": 7.988537296330468e-06, + "loss": 0.6181, + "step": 1335 + }, + { + "epoch": 0.3175659614927502, + "grad_norm": 1.328030117563793, + "learning_rate": 7.985449817912253e-06, + "loss": 0.6073, + "step": 1336 + }, + { + "epoch": 0.3178036605657238, + "grad_norm": 1.3603517103640341, + "learning_rate": 7.982360569429206e-06, + "loss": 0.5906, + "step": 1337 + }, + { + "epoch": 0.3180413596386974, + "grad_norm": 1.3492405526782112, + "learning_rate": 7.979269552712935e-06, + "loss": 0.6611, + "step": 1338 + }, + { + "epoch": 0.318279058711671, + "grad_norm": 1.1852120370940156, + "learning_rate": 7.976176769596095e-06, + "loss": 0.578, + "step": 1339 + }, + { + "epoch": 0.31851675778464467, + "grad_norm": 1.2578126915921488, + "learning_rate": 7.973082221912387e-06, + "loss": 0.6061, + "step": 1340 + }, + { + "epoch": 0.3187544568576183, + "grad_norm": 1.2839072042784052, + "learning_rate": 7.969985911496561e-06, + "loss": 0.6104, + "step": 1341 + }, + { + "epoch": 0.3189921559305919, + "grad_norm": 1.1631005393818197, + "learning_rate": 7.966887840184412e-06, + "loss": 0.5204, + "step": 1342 + }, + { + "epoch": 0.3192298550035655, + "grad_norm": 1.153726385026386, + "learning_rate": 7.963788009812775e-06, + "loss": 0.6296, + "step": 1343 + }, + { + "epoch": 0.3194675540765391, + "grad_norm": 1.1890069679469943, + "learning_rate": 7.960686422219532e-06, + "loss": 0.6171, + "step": 1344 + }, + { + "epoch": 0.3197052531495127, + "grad_norm": 1.3343836783138676, + "learning_rate": 7.957583079243607e-06, + "loss": 0.6503, + "step": 1345 + }, + { + "epoch": 0.3199429522224863, + "grad_norm": 1.3097086221822005, + "learning_rate": 7.954477982724963e-06, + "loss": 0.6244, + "step": 1346 + }, + { + "epoch": 0.32018065129546, + "grad_norm": 1.3961269777291947, + "learning_rate": 7.951371134504599e-06, + "loss": 0.6363, + "step": 1347 + }, + { + "epoch": 0.3204183503684336, + "grad_norm": 1.2334416673248678, + "learning_rate": 7.94826253642456e-06, + "loss": 0.6053, + "step": 1348 + }, + { + "epoch": 0.3206560494414072, + "grad_norm": 1.6434245373709506, + "learning_rate": 7.945152190327926e-06, + "loss": 0.6439, + "step": 1349 + }, + { + "epoch": 0.3208937485143808, + "grad_norm": 1.196493384552022, + "learning_rate": 7.942040098058811e-06, + "loss": 0.6038, + "step": 1350 + }, + { + "epoch": 0.3211314475873544, + "grad_norm": 1.1210977764550953, + "learning_rate": 7.938926261462366e-06, + "loss": 0.6175, + "step": 1351 + }, + { + "epoch": 0.321369146660328, + "grad_norm": 1.32312482626146, + "learning_rate": 7.935810682384777e-06, + "loss": 0.6399, + "step": 1352 + }, + { + "epoch": 0.3216068457333016, + "grad_norm": 1.1694546529076086, + "learning_rate": 7.93269336267326e-06, + "loss": 0.6594, + "step": 1353 + }, + { + "epoch": 0.3218445448062753, + "grad_norm": 1.7938256260251146, + "learning_rate": 7.929574304176067e-06, + "loss": 0.5992, + "step": 1354 + }, + { + "epoch": 0.3220822438792489, + "grad_norm": 1.1415075545142077, + "learning_rate": 7.926453508742478e-06, + "loss": 0.5974, + "step": 1355 + }, + { + "epoch": 0.3223199429522225, + "grad_norm": 1.3023310955030218, + "learning_rate": 7.923330978222802e-06, + "loss": 0.5947, + "step": 1356 + }, + { + "epoch": 0.3225576420251961, + "grad_norm": 1.423700393017611, + "learning_rate": 7.920206714468383e-06, + "loss": 0.668, + "step": 1357 + }, + { + "epoch": 0.3227953410981697, + "grad_norm": 1.2195175387609067, + "learning_rate": 7.917080719331586e-06, + "loss": 0.6479, + "step": 1358 + }, + { + "epoch": 0.3230330401711433, + "grad_norm": 1.209791255955745, + "learning_rate": 7.913952994665805e-06, + "loss": 0.5949, + "step": 1359 + }, + { + "epoch": 0.3232707392441169, + "grad_norm": 1.2494294024196895, + "learning_rate": 7.910823542325459e-06, + "loss": 0.6142, + "step": 1360 + }, + { + "epoch": 0.3235084383170906, + "grad_norm": 1.350561012463484, + "learning_rate": 7.90769236416599e-06, + "loss": 0.6386, + "step": 1361 + }, + { + "epoch": 0.3237461373900642, + "grad_norm": 1.2429652858828513, + "learning_rate": 7.904559462043869e-06, + "loss": 0.6254, + "step": 1362 + }, + { + "epoch": 0.3239838364630378, + "grad_norm": 1.3236730753796173, + "learning_rate": 7.90142483781658e-06, + "loss": 0.6335, + "step": 1363 + }, + { + "epoch": 0.3242215355360114, + "grad_norm": 1.3432421403410193, + "learning_rate": 7.898288493342633e-06, + "loss": 0.6749, + "step": 1364 + }, + { + "epoch": 0.324459234608985, + "grad_norm": 1.2008122552894525, + "learning_rate": 7.895150430481565e-06, + "loss": 0.6828, + "step": 1365 + }, + { + "epoch": 0.3246969336819586, + "grad_norm": 1.2254924362672976, + "learning_rate": 7.892010651093915e-06, + "loss": 0.6095, + "step": 1366 + }, + { + "epoch": 0.32493463275493223, + "grad_norm": 1.4286564477041055, + "learning_rate": 7.888869157041257e-06, + "loss": 0.6019, + "step": 1367 + }, + { + "epoch": 0.3251723318279059, + "grad_norm": 1.1087203569279174, + "learning_rate": 7.88572595018617e-06, + "loss": 0.5993, + "step": 1368 + }, + { + "epoch": 0.3254100309008795, + "grad_norm": 1.2108564250659755, + "learning_rate": 7.882581032392252e-06, + "loss": 0.6421, + "step": 1369 + }, + { + "epoch": 0.3256477299738531, + "grad_norm": 1.3027680672673578, + "learning_rate": 7.879434405524119e-06, + "loss": 0.6269, + "step": 1370 + }, + { + "epoch": 0.3258854290468267, + "grad_norm": 1.2666446779852947, + "learning_rate": 7.876286071447395e-06, + "loss": 0.5956, + "step": 1371 + }, + { + "epoch": 0.3261231281198003, + "grad_norm": 1.3168266807847306, + "learning_rate": 7.873136032028719e-06, + "loss": 0.5777, + "step": 1372 + }, + { + "epoch": 0.32636082719277393, + "grad_norm": 1.4486711196052473, + "learning_rate": 7.869984289135742e-06, + "loss": 0.607, + "step": 1373 + }, + { + "epoch": 0.3265985262657476, + "grad_norm": 1.3816008583113755, + "learning_rate": 7.86683084463712e-06, + "loss": 0.6371, + "step": 1374 + }, + { + "epoch": 0.3268362253387212, + "grad_norm": 1.3253992677438327, + "learning_rate": 7.863675700402527e-06, + "loss": 0.6429, + "step": 1375 + }, + { + "epoch": 0.3270739244116948, + "grad_norm": 1.3853373738445711, + "learning_rate": 7.860518858302634e-06, + "loss": 0.6355, + "step": 1376 + }, + { + "epoch": 0.3273116234846684, + "grad_norm": 1.5702696045132059, + "learning_rate": 7.857360320209126e-06, + "loss": 0.6646, + "step": 1377 + }, + { + "epoch": 0.327549322557642, + "grad_norm": 1.1814127383968338, + "learning_rate": 7.85420008799469e-06, + "loss": 0.627, + "step": 1378 + }, + { + "epoch": 0.3277870216306156, + "grad_norm": 1.2029763554414221, + "learning_rate": 7.85103816353302e-06, + "loss": 0.6675, + "step": 1379 + }, + { + "epoch": 0.32802472070358923, + "grad_norm": 1.2258218109885253, + "learning_rate": 7.847874548698813e-06, + "loss": 0.6267, + "step": 1380 + }, + { + "epoch": 0.3282624197765629, + "grad_norm": 1.1734605173981734, + "learning_rate": 7.844709245367766e-06, + "loss": 0.602, + "step": 1381 + }, + { + "epoch": 0.3285001188495365, + "grad_norm": 1.2922794237854347, + "learning_rate": 7.841542255416579e-06, + "loss": 0.6449, + "step": 1382 + }, + { + "epoch": 0.3287378179225101, + "grad_norm": 1.047278976783117, + "learning_rate": 7.838373580722952e-06, + "loss": 0.6062, + "step": 1383 + }, + { + "epoch": 0.3289755169954837, + "grad_norm": 1.3931378961371914, + "learning_rate": 7.835203223165581e-06, + "loss": 0.6526, + "step": 1384 + }, + { + "epoch": 0.3292132160684573, + "grad_norm": 1.5119236347724851, + "learning_rate": 7.832031184624165e-06, + "loss": 0.6449, + "step": 1385 + }, + { + "epoch": 0.32945091514143093, + "grad_norm": 1.1297121426856196, + "learning_rate": 7.828857466979396e-06, + "loss": 0.6186, + "step": 1386 + }, + { + "epoch": 0.32968861421440454, + "grad_norm": 1.1861439756663321, + "learning_rate": 7.82568207211296e-06, + "loss": 0.6129, + "step": 1387 + }, + { + "epoch": 0.3299263132873782, + "grad_norm": 1.1140985383417783, + "learning_rate": 7.822505001907544e-06, + "loss": 0.5754, + "step": 1388 + }, + { + "epoch": 0.3301640123603518, + "grad_norm": 1.4598123652532033, + "learning_rate": 7.81932625824682e-06, + "loss": 0.6406, + "step": 1389 + }, + { + "epoch": 0.3304017114333254, + "grad_norm": 1.1801483620483901, + "learning_rate": 7.816145843015457e-06, + "loss": 0.6255, + "step": 1390 + }, + { + "epoch": 0.330639410506299, + "grad_norm": 1.2362308172493828, + "learning_rate": 7.812963758099118e-06, + "loss": 0.6509, + "step": 1391 + }, + { + "epoch": 0.33087710957927263, + "grad_norm": 1.6966444324944974, + "learning_rate": 7.809780005384446e-06, + "loss": 0.5942, + "step": 1392 + }, + { + "epoch": 0.33111480865224624, + "grad_norm": 1.3136981487854897, + "learning_rate": 7.806594586759083e-06, + "loss": 0.6169, + "step": 1393 + }, + { + "epoch": 0.33135250772521985, + "grad_norm": 1.2189296063100328, + "learning_rate": 7.803407504111655e-06, + "loss": 0.6257, + "step": 1394 + }, + { + "epoch": 0.3315902067981935, + "grad_norm": 1.3978613154277462, + "learning_rate": 7.800218759331775e-06, + "loss": 0.6367, + "step": 1395 + }, + { + "epoch": 0.3318279058711671, + "grad_norm": 1.2574507729276458, + "learning_rate": 7.797028354310036e-06, + "loss": 0.5953, + "step": 1396 + }, + { + "epoch": 0.3320656049441407, + "grad_norm": 1.3495561193428869, + "learning_rate": 7.793836290938026e-06, + "loss": 0.6071, + "step": 1397 + }, + { + "epoch": 0.33230330401711433, + "grad_norm": 1.0983838631068645, + "learning_rate": 7.790642571108308e-06, + "loss": 0.5998, + "step": 1398 + }, + { + "epoch": 0.33254100309008794, + "grad_norm": 1.3750205744142363, + "learning_rate": 7.787447196714428e-06, + "loss": 0.6181, + "step": 1399 + }, + { + "epoch": 0.33277870216306155, + "grad_norm": 1.5904300859173441, + "learning_rate": 7.784250169650917e-06, + "loss": 0.5632, + "step": 1400 + }, + { + "epoch": 0.33277870216306155, + "eval_loss": 0.6274275183677673, + "eval_runtime": 279.2267, + "eval_samples_per_second": 38.958, + "eval_steps_per_second": 1.218, + "step": 1400 + }, + { + "epoch": 0.33301640123603515, + "grad_norm": 1.3706745245510972, + "learning_rate": 7.781051491813282e-06, + "loss": 0.5986, + "step": 1401 + }, + { + "epoch": 0.3332541003090088, + "grad_norm": 1.2471439156398363, + "learning_rate": 7.777851165098012e-06, + "loss": 0.6084, + "step": 1402 + }, + { + "epoch": 0.3334917993819824, + "grad_norm": 1.1574290491293298, + "learning_rate": 7.774649191402569e-06, + "loss": 0.6144, + "step": 1403 + }, + { + "epoch": 0.33372949845495603, + "grad_norm": 1.3379846237155544, + "learning_rate": 7.771445572625395e-06, + "loss": 0.5893, + "step": 1404 + }, + { + "epoch": 0.33396719752792964, + "grad_norm": 1.2718209352422505, + "learning_rate": 7.768240310665909e-06, + "loss": 0.6192, + "step": 1405 + }, + { + "epoch": 0.33420489660090325, + "grad_norm": 1.2649789694647042, + "learning_rate": 7.765033407424499e-06, + "loss": 0.6403, + "step": 1406 + }, + { + "epoch": 0.33444259567387685, + "grad_norm": 1.194170284528292, + "learning_rate": 7.76182486480253e-06, + "loss": 0.6022, + "step": 1407 + }, + { + "epoch": 0.33468029474685046, + "grad_norm": 1.3140018007196461, + "learning_rate": 7.758614684702338e-06, + "loss": 0.6153, + "step": 1408 + }, + { + "epoch": 0.3349179938198241, + "grad_norm": 1.2645946621255593, + "learning_rate": 7.75540286902723e-06, + "loss": 0.6427, + "step": 1409 + }, + { + "epoch": 0.33515569289279773, + "grad_norm": 1.0930588164756059, + "learning_rate": 7.752189419681481e-06, + "loss": 0.5644, + "step": 1410 + }, + { + "epoch": 0.33539339196577134, + "grad_norm": 1.3529374861621992, + "learning_rate": 7.748974338570337e-06, + "loss": 0.5887, + "step": 1411 + }, + { + "epoch": 0.33563109103874494, + "grad_norm": 1.218202705579788, + "learning_rate": 7.745757627600013e-06, + "loss": 0.619, + "step": 1412 + }, + { + "epoch": 0.33586879011171855, + "grad_norm": 1.194407428216033, + "learning_rate": 7.742539288677682e-06, + "loss": 0.6645, + "step": 1413 + }, + { + "epoch": 0.33610648918469216, + "grad_norm": 1.2318576176092677, + "learning_rate": 7.739319323711495e-06, + "loss": 0.6168, + "step": 1414 + }, + { + "epoch": 0.3363441882576658, + "grad_norm": 1.230151900857714, + "learning_rate": 7.736097734610557e-06, + "loss": 0.6461, + "step": 1415 + }, + { + "epoch": 0.33658188733063943, + "grad_norm": 1.2907326629289178, + "learning_rate": 7.73287452328494e-06, + "loss": 0.6233, + "step": 1416 + }, + { + "epoch": 0.33681958640361304, + "grad_norm": 1.2567099243579207, + "learning_rate": 7.729649691645673e-06, + "loss": 0.5728, + "step": 1417 + }, + { + "epoch": 0.33705728547658664, + "grad_norm": 1.458849091469974, + "learning_rate": 7.726423241604755e-06, + "loss": 0.6287, + "step": 1418 + }, + { + "epoch": 0.33729498454956025, + "grad_norm": 1.282311184413383, + "learning_rate": 7.723195175075136e-06, + "loss": 0.6208, + "step": 1419 + }, + { + "epoch": 0.33753268362253386, + "grad_norm": 1.3302355897027145, + "learning_rate": 7.71996549397073e-06, + "loss": 0.6082, + "step": 1420 + }, + { + "epoch": 0.33777038269550747, + "grad_norm": 1.4538322426678594, + "learning_rate": 7.716734200206403e-06, + "loss": 0.6379, + "step": 1421 + }, + { + "epoch": 0.33800808176848113, + "grad_norm": 1.2158960193177926, + "learning_rate": 7.713501295697983e-06, + "loss": 0.5901, + "step": 1422 + }, + { + "epoch": 0.33824578084145474, + "grad_norm": 1.1863563981323624, + "learning_rate": 7.710266782362248e-06, + "loss": 0.6154, + "step": 1423 + }, + { + "epoch": 0.33848347991442834, + "grad_norm": 1.455747935042879, + "learning_rate": 7.707030662116932e-06, + "loss": 0.584, + "step": 1424 + }, + { + "epoch": 0.33872117898740195, + "grad_norm": 1.3368709418627875, + "learning_rate": 7.703792936880722e-06, + "loss": 0.6061, + "step": 1425 + }, + { + "epoch": 0.33895887806037556, + "grad_norm": 1.1499607328436587, + "learning_rate": 7.700553608573257e-06, + "loss": 0.6383, + "step": 1426 + }, + { + "epoch": 0.33919657713334916, + "grad_norm": 1.3038366416402367, + "learning_rate": 7.697312679115126e-06, + "loss": 0.592, + "step": 1427 + }, + { + "epoch": 0.33943427620632277, + "grad_norm": 1.96902718860751, + "learning_rate": 7.694070150427864e-06, + "loss": 0.6675, + "step": 1428 + }, + { + "epoch": 0.33967197527929643, + "grad_norm": 1.4283171826272125, + "learning_rate": 7.69082602443396e-06, + "loss": 0.7066, + "step": 1429 + }, + { + "epoch": 0.33990967435227004, + "grad_norm": 1.3498992453196497, + "learning_rate": 7.687580303056844e-06, + "loss": 0.6127, + "step": 1430 + }, + { + "epoch": 0.34014737342524365, + "grad_norm": 1.145291878264642, + "learning_rate": 7.684332988220901e-06, + "loss": 0.5992, + "step": 1431 + }, + { + "epoch": 0.34038507249821726, + "grad_norm": 1.2655098364637234, + "learning_rate": 7.68108408185145e-06, + "loss": 0.614, + "step": 1432 + }, + { + "epoch": 0.34062277157119086, + "grad_norm": 1.2417467124045936, + "learning_rate": 7.67783358587476e-06, + "loss": 0.6249, + "step": 1433 + }, + { + "epoch": 0.34086047064416447, + "grad_norm": 1.4487182612600435, + "learning_rate": 7.67458150221804e-06, + "loss": 0.6265, + "step": 1434 + }, + { + "epoch": 0.3410981697171381, + "grad_norm": 1.4054567379380285, + "learning_rate": 7.671327832809442e-06, + "loss": 0.5802, + "step": 1435 + }, + { + "epoch": 0.34133586879011174, + "grad_norm": 1.2275721238244477, + "learning_rate": 7.66807257957806e-06, + "loss": 0.6147, + "step": 1436 + }, + { + "epoch": 0.34157356786308535, + "grad_norm": 1.320966041197093, + "learning_rate": 7.664815744453918e-06, + "loss": 0.6169, + "step": 1437 + }, + { + "epoch": 0.34181126693605896, + "grad_norm": 1.2294594627420763, + "learning_rate": 7.66155732936799e-06, + "loss": 0.6375, + "step": 1438 + }, + { + "epoch": 0.34204896600903256, + "grad_norm": 1.2618135758703612, + "learning_rate": 7.658297336252181e-06, + "loss": 0.6735, + "step": 1439 + }, + { + "epoch": 0.34228666508200617, + "grad_norm": 1.2809939998281623, + "learning_rate": 7.65503576703933e-06, + "loss": 0.6632, + "step": 1440 + }, + { + "epoch": 0.3425243641549798, + "grad_norm": 1.1534424867956883, + "learning_rate": 7.651772623663212e-06, + "loss": 0.6733, + "step": 1441 + }, + { + "epoch": 0.3427620632279534, + "grad_norm": 1.1181916760066815, + "learning_rate": 7.648507908058536e-06, + "loss": 0.6005, + "step": 1442 + }, + { + "epoch": 0.34299976230092705, + "grad_norm": 1.2600094533842598, + "learning_rate": 7.645241622160944e-06, + "loss": 0.6111, + "step": 1443 + }, + { + "epoch": 0.34323746137390065, + "grad_norm": 1.219461951838587, + "learning_rate": 7.641973767907006e-06, + "loss": 0.6346, + "step": 1444 + }, + { + "epoch": 0.34347516044687426, + "grad_norm": 1.299098687315041, + "learning_rate": 7.638704347234224e-06, + "loss": 0.5807, + "step": 1445 + }, + { + "epoch": 0.34371285951984787, + "grad_norm": 1.1627385216603177, + "learning_rate": 7.63543336208103e-06, + "loss": 0.5829, + "step": 1446 + }, + { + "epoch": 0.3439505585928215, + "grad_norm": 1.8691015271827454, + "learning_rate": 7.63216081438678e-06, + "loss": 0.6545, + "step": 1447 + }, + { + "epoch": 0.3441882576657951, + "grad_norm": 1.207519533943871, + "learning_rate": 7.628886706091758e-06, + "loss": 0.6494, + "step": 1448 + }, + { + "epoch": 0.34442595673876875, + "grad_norm": 1.1908550395565431, + "learning_rate": 7.625611039137176e-06, + "loss": 0.6203, + "step": 1449 + }, + { + "epoch": 0.34466365581174235, + "grad_norm": 1.2803748960076193, + "learning_rate": 7.622333815465166e-06, + "loss": 0.5969, + "step": 1450 + }, + { + "epoch": 0.34490135488471596, + "grad_norm": 1.8339247119429252, + "learning_rate": 7.619055037018785e-06, + "loss": 0.6085, + "step": 1451 + }, + { + "epoch": 0.34513905395768957, + "grad_norm": 1.2139167519973701, + "learning_rate": 7.615774705742012e-06, + "loss": 0.5842, + "step": 1452 + }, + { + "epoch": 0.3453767530306632, + "grad_norm": 1.6865156216545618, + "learning_rate": 7.612492823579744e-06, + "loss": 0.6231, + "step": 1453 + }, + { + "epoch": 0.3456144521036368, + "grad_norm": 1.2725699304550304, + "learning_rate": 7.609209392477803e-06, + "loss": 0.6519, + "step": 1454 + }, + { + "epoch": 0.3458521511766104, + "grad_norm": 1.2641122286195516, + "learning_rate": 7.605924414382926e-06, + "loss": 0.6041, + "step": 1455 + }, + { + "epoch": 0.34608985024958405, + "grad_norm": 1.2326709985394138, + "learning_rate": 7.602637891242765e-06, + "loss": 0.6227, + "step": 1456 + }, + { + "epoch": 0.34632754932255766, + "grad_norm": 1.188574926530305, + "learning_rate": 7.599349825005892e-06, + "loss": 0.6048, + "step": 1457 + }, + { + "epoch": 0.34656524839553127, + "grad_norm": 1.1444687702050254, + "learning_rate": 7.59606021762179e-06, + "loss": 0.577, + "step": 1458 + }, + { + "epoch": 0.3468029474685049, + "grad_norm": 1.3020702446649142, + "learning_rate": 7.5927690710408606e-06, + "loss": 0.6258, + "step": 1459 + }, + { + "epoch": 0.3470406465414785, + "grad_norm": 2.123017651995483, + "learning_rate": 7.589476387214414e-06, + "loss": 0.616, + "step": 1460 + }, + { + "epoch": 0.3472783456144521, + "grad_norm": 1.2167554953441788, + "learning_rate": 7.586182168094675e-06, + "loss": 0.65, + "step": 1461 + }, + { + "epoch": 0.3475160446874257, + "grad_norm": 1.2049280784819254, + "learning_rate": 7.5828864156347735e-06, + "loss": 0.5599, + "step": 1462 + }, + { + "epoch": 0.34775374376039936, + "grad_norm": 1.3917891266743012, + "learning_rate": 7.579589131788756e-06, + "loss": 0.589, + "step": 1463 + }, + { + "epoch": 0.34799144283337297, + "grad_norm": 1.0939173226231835, + "learning_rate": 7.57629031851157e-06, + "loss": 0.597, + "step": 1464 + }, + { + "epoch": 0.3482291419063466, + "grad_norm": 1.2322932427768616, + "learning_rate": 7.572989977759073e-06, + "loss": 0.6009, + "step": 1465 + }, + { + "epoch": 0.3484668409793202, + "grad_norm": 1.7863512727543354, + "learning_rate": 7.569688111488029e-06, + "loss": 0.5894, + "step": 1466 + }, + { + "epoch": 0.3487045400522938, + "grad_norm": 1.3021564832011114, + "learning_rate": 7.566384721656103e-06, + "loss": 0.6232, + "step": 1467 + }, + { + "epoch": 0.3489422391252674, + "grad_norm": 1.3082763411955132, + "learning_rate": 7.563079810221869e-06, + "loss": 0.6323, + "step": 1468 + }, + { + "epoch": 0.349179938198241, + "grad_norm": 1.4051250505609907, + "learning_rate": 7.559773379144797e-06, + "loss": 0.6245, + "step": 1469 + }, + { + "epoch": 0.34941763727121466, + "grad_norm": 1.3922396261940628, + "learning_rate": 7.55646543038526e-06, + "loss": 0.5725, + "step": 1470 + }, + { + "epoch": 0.3496553363441883, + "grad_norm": 1.207493231317218, + "learning_rate": 7.553155965904535e-06, + "loss": 0.62, + "step": 1471 + }, + { + "epoch": 0.3498930354171619, + "grad_norm": 1.3378102187732477, + "learning_rate": 7.549844987664792e-06, + "loss": 0.5813, + "step": 1472 + }, + { + "epoch": 0.3501307344901355, + "grad_norm": 1.1765559923411713, + "learning_rate": 7.546532497629101e-06, + "loss": 0.5453, + "step": 1473 + }, + { + "epoch": 0.3503684335631091, + "grad_norm": 1.4082945112285195, + "learning_rate": 7.543218497761427e-06, + "loss": 0.6254, + "step": 1474 + }, + { + "epoch": 0.3506061326360827, + "grad_norm": 1.3868930744001304, + "learning_rate": 7.539902990026634e-06, + "loss": 0.6372, + "step": 1475 + }, + { + "epoch": 0.3508438317090563, + "grad_norm": 1.2690724745966777, + "learning_rate": 7.536585976390474e-06, + "loss": 0.5909, + "step": 1476 + }, + { + "epoch": 0.35108153078202997, + "grad_norm": 1.355636557597816, + "learning_rate": 7.533267458819597e-06, + "loss": 0.571, + "step": 1477 + }, + { + "epoch": 0.3513192298550036, + "grad_norm": 1.314193029952112, + "learning_rate": 7.529947439281542e-06, + "loss": 0.6453, + "step": 1478 + }, + { + "epoch": 0.3515569289279772, + "grad_norm": 1.2851014960880245, + "learning_rate": 7.526625919744741e-06, + "loss": 0.6454, + "step": 1479 + }, + { + "epoch": 0.3517946280009508, + "grad_norm": 1.5119448852799269, + "learning_rate": 7.523302902178511e-06, + "loss": 0.6538, + "step": 1480 + }, + { + "epoch": 0.3520323270739244, + "grad_norm": 1.1314192842788457, + "learning_rate": 7.519978388553062e-06, + "loss": 0.5955, + "step": 1481 + }, + { + "epoch": 0.352270026146898, + "grad_norm": 1.156587082015866, + "learning_rate": 7.516652380839488e-06, + "loss": 0.6207, + "step": 1482 + }, + { + "epoch": 0.3525077252198716, + "grad_norm": 1.1966051368841975, + "learning_rate": 7.513324881009769e-06, + "loss": 0.6154, + "step": 1483 + }, + { + "epoch": 0.3527454242928453, + "grad_norm": 1.2958725546150647, + "learning_rate": 7.50999589103677e-06, + "loss": 0.6325, + "step": 1484 + }, + { + "epoch": 0.3529831233658189, + "grad_norm": 1.2099461682322017, + "learning_rate": 7.506665412894243e-06, + "loss": 0.639, + "step": 1485 + }, + { + "epoch": 0.3532208224387925, + "grad_norm": 1.2787459578600893, + "learning_rate": 7.503333448556816e-06, + "loss": 0.6459, + "step": 1486 + }, + { + "epoch": 0.3534585215117661, + "grad_norm": 1.1434017696021395, + "learning_rate": 7.500000000000001e-06, + "loss": 0.6483, + "step": 1487 + }, + { + "epoch": 0.3536962205847397, + "grad_norm": 1.4538139306650957, + "learning_rate": 7.496665069200192e-06, + "loss": 0.5499, + "step": 1488 + }, + { + "epoch": 0.3539339196577133, + "grad_norm": 1.2702446595257346, + "learning_rate": 7.49332865813466e-06, + "loss": 0.5969, + "step": 1489 + }, + { + "epoch": 0.354171618730687, + "grad_norm": 1.0768941349795227, + "learning_rate": 7.489990768781553e-06, + "loss": 0.6447, + "step": 1490 + }, + { + "epoch": 0.3544093178036606, + "grad_norm": 1.237352856665914, + "learning_rate": 7.486651403119898e-06, + "loss": 0.6352, + "step": 1491 + }, + { + "epoch": 0.3546470168766342, + "grad_norm": 1.2034453729249124, + "learning_rate": 7.483310563129592e-06, + "loss": 0.6273, + "step": 1492 + }, + { + "epoch": 0.3548847159496078, + "grad_norm": 1.1853680891453284, + "learning_rate": 7.4799682507914136e-06, + "loss": 0.6105, + "step": 1493 + }, + { + "epoch": 0.3551224150225814, + "grad_norm": 1.1995536182687923, + "learning_rate": 7.476624468087008e-06, + "loss": 0.6061, + "step": 1494 + }, + { + "epoch": 0.355360114095555, + "grad_norm": 1.3452927067970164, + "learning_rate": 7.473279216998896e-06, + "loss": 0.6002, + "step": 1495 + }, + { + "epoch": 0.3555978131685286, + "grad_norm": 1.361573998294133, + "learning_rate": 7.469932499510465e-06, + "loss": 0.6542, + "step": 1496 + }, + { + "epoch": 0.3558355122415023, + "grad_norm": 1.41775412515146, + "learning_rate": 7.466584317605978e-06, + "loss": 0.598, + "step": 1497 + }, + { + "epoch": 0.3560732113144759, + "grad_norm": 1.2144638439851962, + "learning_rate": 7.46323467327056e-06, + "loss": 0.6109, + "step": 1498 + }, + { + "epoch": 0.3563109103874495, + "grad_norm": 1.2489662138324993, + "learning_rate": 7.459883568490207e-06, + "loss": 0.6427, + "step": 1499 + }, + { + "epoch": 0.3565486094604231, + "grad_norm": 1.2366337451628406, + "learning_rate": 7.456531005251779e-06, + "loss": 0.5939, + "step": 1500 + }, + { + "epoch": 0.3565486094604231, + "eval_loss": 0.6231275796890259, + "eval_runtime": 284.3133, + "eval_samples_per_second": 38.261, + "eval_steps_per_second": 1.196, + "step": 1500 + }, + { + "epoch": 0.3567863085333967, + "grad_norm": 1.2438170717518602, + "learning_rate": 7.453176985543002e-06, + "loss": 0.6328, + "step": 1501 + }, + { + "epoch": 0.3570240076063703, + "grad_norm": 1.2251887741050849, + "learning_rate": 7.449821511352465e-06, + "loss": 0.6231, + "step": 1502 + }, + { + "epoch": 0.3572617066793439, + "grad_norm": 1.6507680007363423, + "learning_rate": 7.4464645846696186e-06, + "loss": 0.5833, + "step": 1503 + }, + { + "epoch": 0.3574994057523176, + "grad_norm": 1.1974254190305775, + "learning_rate": 7.443106207484776e-06, + "loss": 0.5927, + "step": 1504 + }, + { + "epoch": 0.3577371048252912, + "grad_norm": 1.2546449605357763, + "learning_rate": 7.439746381789108e-06, + "loss": 0.6253, + "step": 1505 + }, + { + "epoch": 0.3579748038982648, + "grad_norm": 1.316487375385313, + "learning_rate": 7.436385109574647e-06, + "loss": 0.6628, + "step": 1506 + }, + { + "epoch": 0.3582125029712384, + "grad_norm": 1.2631421255965267, + "learning_rate": 7.4330223928342814e-06, + "loss": 0.6956, + "step": 1507 + }, + { + "epoch": 0.358450202044212, + "grad_norm": 1.3887993262154423, + "learning_rate": 7.429658233561759e-06, + "loss": 0.6221, + "step": 1508 + }, + { + "epoch": 0.3586879011171856, + "grad_norm": 1.1583472365384284, + "learning_rate": 7.426292633751676e-06, + "loss": 0.5996, + "step": 1509 + }, + { + "epoch": 0.35892560019015923, + "grad_norm": 1.3089183270706124, + "learning_rate": 7.422925595399491e-06, + "loss": 0.6157, + "step": 1510 + }, + { + "epoch": 0.3591632992631329, + "grad_norm": 1.208325138407309, + "learning_rate": 7.419557120501508e-06, + "loss": 0.664, + "step": 1511 + }, + { + "epoch": 0.3594009983361065, + "grad_norm": 1.2401254464464435, + "learning_rate": 7.416187211054891e-06, + "loss": 0.5927, + "step": 1512 + }, + { + "epoch": 0.3596386974090801, + "grad_norm": 1.2177982316824267, + "learning_rate": 7.412815869057644e-06, + "loss": 0.5941, + "step": 1513 + }, + { + "epoch": 0.3598763964820537, + "grad_norm": 1.2654804631808385, + "learning_rate": 7.4094430965086325e-06, + "loss": 0.6163, + "step": 1514 + }, + { + "epoch": 0.3601140955550273, + "grad_norm": 1.184907660293741, + "learning_rate": 7.406068895407559e-06, + "loss": 0.5937, + "step": 1515 + }, + { + "epoch": 0.36035179462800093, + "grad_norm": 1.4878024966678596, + "learning_rate": 7.402693267754981e-06, + "loss": 0.5843, + "step": 1516 + }, + { + "epoch": 0.36058949370097454, + "grad_norm": 1.2126271733564062, + "learning_rate": 7.399316215552296e-06, + "loss": 0.6102, + "step": 1517 + }, + { + "epoch": 0.3608271927739482, + "grad_norm": 1.2537949935420396, + "learning_rate": 7.3959377408017505e-06, + "loss": 0.5922, + "step": 1518 + }, + { + "epoch": 0.3610648918469218, + "grad_norm": 1.2594398900133135, + "learning_rate": 7.392557845506433e-06, + "loss": 0.5851, + "step": 1519 + }, + { + "epoch": 0.3613025909198954, + "grad_norm": 1.0842578092271564, + "learning_rate": 7.389176531670273e-06, + "loss": 0.5981, + "step": 1520 + }, + { + "epoch": 0.361540289992869, + "grad_norm": 1.3138559320364764, + "learning_rate": 7.3857938012980425e-06, + "loss": 0.6037, + "step": 1521 + }, + { + "epoch": 0.36177798906584263, + "grad_norm": 1.1473059539294224, + "learning_rate": 7.382409656395353e-06, + "loss": 0.5849, + "step": 1522 + }, + { + "epoch": 0.36201568813881624, + "grad_norm": 2.0690971463669636, + "learning_rate": 7.379024098968653e-06, + "loss": 0.6206, + "step": 1523 + }, + { + "epoch": 0.3622533872117899, + "grad_norm": 1.3421571573729283, + "learning_rate": 7.3756371310252325e-06, + "loss": 0.6341, + "step": 1524 + }, + { + "epoch": 0.3624910862847635, + "grad_norm": 1.4004867722805712, + "learning_rate": 7.372248754573213e-06, + "loss": 0.5875, + "step": 1525 + }, + { + "epoch": 0.3627287853577371, + "grad_norm": 1.39963252728955, + "learning_rate": 7.3688589716215555e-06, + "loss": 0.5819, + "step": 1526 + }, + { + "epoch": 0.3629664844307107, + "grad_norm": 1.2190211581272317, + "learning_rate": 7.365467784180051e-06, + "loss": 0.5746, + "step": 1527 + }, + { + "epoch": 0.36320418350368433, + "grad_norm": 1.0980829664608636, + "learning_rate": 7.362075194259326e-06, + "loss": 0.6528, + "step": 1528 + }, + { + "epoch": 0.36344188257665794, + "grad_norm": 1.2197743364006775, + "learning_rate": 7.358681203870836e-06, + "loss": 0.551, + "step": 1529 + }, + { + "epoch": 0.36367958164963154, + "grad_norm": 1.4257169972041883, + "learning_rate": 7.3552858150268726e-06, + "loss": 0.6011, + "step": 1530 + }, + { + "epoch": 0.3639172807226052, + "grad_norm": 1.973643378399761, + "learning_rate": 7.351889029740548e-06, + "loss": 0.6066, + "step": 1531 + }, + { + "epoch": 0.3641549797955788, + "grad_norm": 1.344504041154379, + "learning_rate": 7.348490850025808e-06, + "loss": 0.6158, + "step": 1532 + }, + { + "epoch": 0.3643926788685524, + "grad_norm": 1.7710688509519026, + "learning_rate": 7.3450912778974236e-06, + "loss": 0.6091, + "step": 1533 + }, + { + "epoch": 0.36463037794152603, + "grad_norm": 1.3093516731384451, + "learning_rate": 7.341690315370993e-06, + "loss": 0.6694, + "step": 1534 + }, + { + "epoch": 0.36486807701449964, + "grad_norm": 1.4921239803682163, + "learning_rate": 7.3382879644629345e-06, + "loss": 0.6443, + "step": 1535 + }, + { + "epoch": 0.36510577608747324, + "grad_norm": 1.2507033962003804, + "learning_rate": 7.334884227190496e-06, + "loss": 0.6643, + "step": 1536 + }, + { + "epoch": 0.36534347516044685, + "grad_norm": 1.1741864278078802, + "learning_rate": 7.33147910557174e-06, + "loss": 0.5804, + "step": 1537 + }, + { + "epoch": 0.3655811742334205, + "grad_norm": 1.323615920492878, + "learning_rate": 7.328072601625558e-06, + "loss": 0.6111, + "step": 1538 + }, + { + "epoch": 0.3658188733063941, + "grad_norm": 1.4701656122774496, + "learning_rate": 7.324664717371652e-06, + "loss": 0.6047, + "step": 1539 + }, + { + "epoch": 0.36605657237936773, + "grad_norm": 1.2114364922562202, + "learning_rate": 7.3212554548305515e-06, + "loss": 0.6449, + "step": 1540 + }, + { + "epoch": 0.36629427145234134, + "grad_norm": 1.2542798977523444, + "learning_rate": 7.3178448160235945e-06, + "loss": 0.6088, + "step": 1541 + }, + { + "epoch": 0.36653197052531494, + "grad_norm": 1.486888766001767, + "learning_rate": 7.314432802972943e-06, + "loss": 0.6158, + "step": 1542 + }, + { + "epoch": 0.36676966959828855, + "grad_norm": 1.1584652084779403, + "learning_rate": 7.311019417701567e-06, + "loss": 0.6502, + "step": 1543 + }, + { + "epoch": 0.36700736867126216, + "grad_norm": 1.143388061941866, + "learning_rate": 7.3076046622332546e-06, + "loss": 0.5859, + "step": 1544 + }, + { + "epoch": 0.3672450677442358, + "grad_norm": 1.226236314119809, + "learning_rate": 7.304188538592604e-06, + "loss": 0.5566, + "step": 1545 + }, + { + "epoch": 0.3674827668172094, + "grad_norm": 1.329098658300653, + "learning_rate": 7.300771048805026e-06, + "loss": 0.5693, + "step": 1546 + }, + { + "epoch": 0.36772046589018303, + "grad_norm": 1.184429762464481, + "learning_rate": 7.297352194896738e-06, + "loss": 0.6142, + "step": 1547 + }, + { + "epoch": 0.36795816496315664, + "grad_norm": 1.2625365255072603, + "learning_rate": 7.293931978894775e-06, + "loss": 0.6293, + "step": 1548 + }, + { + "epoch": 0.36819586403613025, + "grad_norm": 1.1485709572304768, + "learning_rate": 7.290510402826967e-06, + "loss": 0.6345, + "step": 1549 + }, + { + "epoch": 0.36843356310910386, + "grad_norm": 1.2494854145370065, + "learning_rate": 7.287087468721962e-06, + "loss": 0.6138, + "step": 1550 + }, + { + "epoch": 0.36867126218207746, + "grad_norm": 1.2955599469351353, + "learning_rate": 7.283663178609204e-06, + "loss": 0.5899, + "step": 1551 + }, + { + "epoch": 0.3689089612550511, + "grad_norm": 1.2013287364734493, + "learning_rate": 7.280237534518948e-06, + "loss": 0.6122, + "step": 1552 + }, + { + "epoch": 0.36914666032802473, + "grad_norm": 1.2140579800753575, + "learning_rate": 7.276810538482245e-06, + "loss": 0.6249, + "step": 1553 + }, + { + "epoch": 0.36938435940099834, + "grad_norm": 1.1977708061216374, + "learning_rate": 7.273382192530957e-06, + "loss": 0.6115, + "step": 1554 + }, + { + "epoch": 0.36962205847397195, + "grad_norm": 1.2599479579368467, + "learning_rate": 7.269952498697734e-06, + "loss": 0.6171, + "step": 1555 + }, + { + "epoch": 0.36985975754694556, + "grad_norm": 1.2793663730675582, + "learning_rate": 7.266521459016038e-06, + "loss": 0.6093, + "step": 1556 + }, + { + "epoch": 0.37009745661991916, + "grad_norm": 1.2794501491895558, + "learning_rate": 7.26308907552012e-06, + "loss": 0.6541, + "step": 1557 + }, + { + "epoch": 0.37033515569289277, + "grad_norm": 1.3531345302547602, + "learning_rate": 7.259655350245031e-06, + "loss": 0.6178, + "step": 1558 + }, + { + "epoch": 0.37057285476586643, + "grad_norm": 1.4442858786138135, + "learning_rate": 7.256220285226615e-06, + "loss": 0.6127, + "step": 1559 + }, + { + "epoch": 0.37081055383884004, + "grad_norm": 1.4361456469568228, + "learning_rate": 7.2527838825015155e-06, + "loss": 0.5795, + "step": 1560 + }, + { + "epoch": 0.37104825291181365, + "grad_norm": 1.4015806339081442, + "learning_rate": 7.249346144107165e-06, + "loss": 0.6013, + "step": 1561 + }, + { + "epoch": 0.37128595198478725, + "grad_norm": 1.0873796310583326, + "learning_rate": 7.245907072081787e-06, + "loss": 0.6168, + "step": 1562 + }, + { + "epoch": 0.37152365105776086, + "grad_norm": 1.319730391273647, + "learning_rate": 7.242466668464402e-06, + "loss": 0.5842, + "step": 1563 + }, + { + "epoch": 0.37176135013073447, + "grad_norm": 1.3479289087920103, + "learning_rate": 7.2390249352948115e-06, + "loss": 0.6028, + "step": 1564 + }, + { + "epoch": 0.37199904920370813, + "grad_norm": 1.1368732564396729, + "learning_rate": 7.235581874613612e-06, + "loss": 0.6056, + "step": 1565 + }, + { + "epoch": 0.37223674827668174, + "grad_norm": 1.290791796894279, + "learning_rate": 7.232137488462183e-06, + "loss": 0.6532, + "step": 1566 + }, + { + "epoch": 0.37247444734965535, + "grad_norm": 1.2592118736202371, + "learning_rate": 7.2286917788826926e-06, + "loss": 0.5983, + "step": 1567 + }, + { + "epoch": 0.37271214642262895, + "grad_norm": 1.8060120666607506, + "learning_rate": 7.22524474791809e-06, + "loss": 0.601, + "step": 1568 + }, + { + "epoch": 0.37294984549560256, + "grad_norm": 1.1504584941626215, + "learning_rate": 7.221796397612114e-06, + "loss": 0.6044, + "step": 1569 + }, + { + "epoch": 0.37318754456857617, + "grad_norm": 1.1976593053762175, + "learning_rate": 7.218346730009279e-06, + "loss": 0.615, + "step": 1570 + }, + { + "epoch": 0.3734252436415498, + "grad_norm": 1.1559219172202062, + "learning_rate": 7.214895747154884e-06, + "loss": 0.5969, + "step": 1571 + }, + { + "epoch": 0.37366294271452344, + "grad_norm": 1.1954922039130218, + "learning_rate": 7.211443451095007e-06, + "loss": 0.5847, + "step": 1572 + }, + { + "epoch": 0.37390064178749705, + "grad_norm": 1.5000216268696476, + "learning_rate": 7.207989843876505e-06, + "loss": 0.5933, + "step": 1573 + }, + { + "epoch": 0.37413834086047065, + "grad_norm": 1.228310450971522, + "learning_rate": 7.204534927547011e-06, + "loss": 0.5893, + "step": 1574 + }, + { + "epoch": 0.37437603993344426, + "grad_norm": 1.2307854199236894, + "learning_rate": 7.201078704154938e-06, + "loss": 0.6309, + "step": 1575 + }, + { + "epoch": 0.37461373900641787, + "grad_norm": 1.2076887662117097, + "learning_rate": 7.197621175749467e-06, + "loss": 0.6165, + "step": 1576 + }, + { + "epoch": 0.3748514380793915, + "grad_norm": 1.4253705276837947, + "learning_rate": 7.194162344380561e-06, + "loss": 0.5736, + "step": 1577 + }, + { + "epoch": 0.3750891371523651, + "grad_norm": 1.3001735360619235, + "learning_rate": 7.1907022120989474e-06, + "loss": 0.5853, + "step": 1578 + }, + { + "epoch": 0.37532683622533874, + "grad_norm": 1.3840179401293142, + "learning_rate": 7.187240780956133e-06, + "loss": 0.6411, + "step": 1579 + }, + { + "epoch": 0.37556453529831235, + "grad_norm": 1.2365855858722528, + "learning_rate": 7.183778053004386e-06, + "loss": 0.692, + "step": 1580 + }, + { + "epoch": 0.37580223437128596, + "grad_norm": 1.2769999057172725, + "learning_rate": 7.180314030296754e-06, + "loss": 0.6136, + "step": 1581 + }, + { + "epoch": 0.37603993344425957, + "grad_norm": 1.1814629028033223, + "learning_rate": 7.176848714887042e-06, + "loss": 0.584, + "step": 1582 + }, + { + "epoch": 0.3762776325172332, + "grad_norm": 1.1488905383643901, + "learning_rate": 7.173382108829826e-06, + "loss": 0.5974, + "step": 1583 + }, + { + "epoch": 0.3765153315902068, + "grad_norm": 1.268128560880298, + "learning_rate": 7.1699142141804475e-06, + "loss": 0.5914, + "step": 1584 + }, + { + "epoch": 0.3767530306631804, + "grad_norm": 1.2337892544993625, + "learning_rate": 7.166445032995013e-06, + "loss": 0.6056, + "step": 1585 + }, + { + "epoch": 0.37699072973615405, + "grad_norm": 1.200249349587499, + "learning_rate": 7.16297456733039e-06, + "loss": 0.5805, + "step": 1586 + }, + { + "epoch": 0.37722842880912766, + "grad_norm": 1.2617210749376133, + "learning_rate": 7.159502819244206e-06, + "loss": 0.6512, + "step": 1587 + }, + { + "epoch": 0.37746612788210127, + "grad_norm": 1.2597383010575467, + "learning_rate": 7.156029790794852e-06, + "loss": 0.6073, + "step": 1588 + }, + { + "epoch": 0.3777038269550749, + "grad_norm": 1.2966643008132208, + "learning_rate": 7.1525554840414765e-06, + "loss": 0.6144, + "step": 1589 + }, + { + "epoch": 0.3779415260280485, + "grad_norm": 1.1899440229692086, + "learning_rate": 7.149079901043987e-06, + "loss": 0.6117, + "step": 1590 + }, + { + "epoch": 0.3781792251010221, + "grad_norm": 1.2799476550436775, + "learning_rate": 7.145603043863045e-06, + "loss": 0.6389, + "step": 1591 + }, + { + "epoch": 0.3784169241739957, + "grad_norm": 1.1818968454540661, + "learning_rate": 7.142124914560073e-06, + "loss": 0.6299, + "step": 1592 + }, + { + "epoch": 0.37865462324696936, + "grad_norm": 1.386902020771797, + "learning_rate": 7.13864551519724e-06, + "loss": 0.5933, + "step": 1593 + }, + { + "epoch": 0.37889232231994296, + "grad_norm": 1.3359346306826154, + "learning_rate": 7.135164847837475e-06, + "loss": 0.6206, + "step": 1594 + }, + { + "epoch": 0.37913002139291657, + "grad_norm": 1.2768280751300933, + "learning_rate": 7.131682914544454e-06, + "loss": 0.6166, + "step": 1595 + }, + { + "epoch": 0.3793677204658902, + "grad_norm": 1.3236541587740813, + "learning_rate": 7.128199717382607e-06, + "loss": 0.6504, + "step": 1596 + }, + { + "epoch": 0.3796054195388638, + "grad_norm": 1.2604074163913956, + "learning_rate": 7.124715258417111e-06, + "loss": 0.643, + "step": 1597 + }, + { + "epoch": 0.3798431186118374, + "grad_norm": 1.2797589977831112, + "learning_rate": 7.121229539713891e-06, + "loss": 0.588, + "step": 1598 + }, + { + "epoch": 0.38008081768481106, + "grad_norm": 1.3970715818380388, + "learning_rate": 7.117742563339622e-06, + "loss": 0.632, + "step": 1599 + }, + { + "epoch": 0.38031851675778466, + "grad_norm": 1.2565425112187032, + "learning_rate": 7.1142543313617214e-06, + "loss": 0.6296, + "step": 1600 + }, + { + "epoch": 0.38031851675778466, + "eval_loss": 0.6184806823730469, + "eval_runtime": 284.6865, + "eval_samples_per_second": 38.21, + "eval_steps_per_second": 1.194, + "step": 1600 + }, + { + "epoch": 0.38055621583075827, + "grad_norm": 3.701927601401408, + "learning_rate": 7.1107648458483535e-06, + "loss": 0.5875, + "step": 1601 + }, + { + "epoch": 0.3807939149037319, + "grad_norm": 1.7047506521797124, + "learning_rate": 7.107274108868422e-06, + "loss": 0.622, + "step": 1602 + }, + { + "epoch": 0.3810316139767055, + "grad_norm": 1.1820188186987284, + "learning_rate": 7.103782122491577e-06, + "loss": 0.6209, + "step": 1603 + }, + { + "epoch": 0.3812693130496791, + "grad_norm": 1.1630571345344443, + "learning_rate": 7.100288888788206e-06, + "loss": 0.6407, + "step": 1604 + }, + { + "epoch": 0.3815070121226527, + "grad_norm": 1.616416600562063, + "learning_rate": 7.096794409829439e-06, + "loss": 0.6815, + "step": 1605 + }, + { + "epoch": 0.38174471119562636, + "grad_norm": 1.5072253752715288, + "learning_rate": 7.093298687687141e-06, + "loss": 0.6314, + "step": 1606 + }, + { + "epoch": 0.38198241026859997, + "grad_norm": 1.3940569283688728, + "learning_rate": 7.089801724433918e-06, + "loss": 0.6171, + "step": 1607 + }, + { + "epoch": 0.3822201093415736, + "grad_norm": 1.4644885099606646, + "learning_rate": 7.086303522143107e-06, + "loss": 0.6439, + "step": 1608 + }, + { + "epoch": 0.3824578084145472, + "grad_norm": 1.3233750178416055, + "learning_rate": 7.082804082888787e-06, + "loss": 0.6316, + "step": 1609 + }, + { + "epoch": 0.3826955074875208, + "grad_norm": 1.3705051849097478, + "learning_rate": 7.07930340874576e-06, + "loss": 0.5996, + "step": 1610 + }, + { + "epoch": 0.3829332065604944, + "grad_norm": 1.3298803125282255, + "learning_rate": 7.07580150178957e-06, + "loss": 0.6529, + "step": 1611 + }, + { + "epoch": 0.383170905633468, + "grad_norm": 1.2701021782333373, + "learning_rate": 7.072298364096486e-06, + "loss": 0.6178, + "step": 1612 + }, + { + "epoch": 0.38340860470644167, + "grad_norm": 1.2738808035996427, + "learning_rate": 7.0687939977435096e-06, + "loss": 0.6166, + "step": 1613 + }, + { + "epoch": 0.3836463037794153, + "grad_norm": 1.3409118627988554, + "learning_rate": 7.065288404808368e-06, + "loss": 0.587, + "step": 1614 + }, + { + "epoch": 0.3838840028523889, + "grad_norm": 1.1718689797934538, + "learning_rate": 7.061781587369518e-06, + "loss": 0.632, + "step": 1615 + }, + { + "epoch": 0.3841217019253625, + "grad_norm": 1.9554088114279746, + "learning_rate": 7.058273547506143e-06, + "loss": 0.6074, + "step": 1616 + }, + { + "epoch": 0.3843594009983361, + "grad_norm": 1.3901778872420052, + "learning_rate": 7.05476428729815e-06, + "loss": 0.6426, + "step": 1617 + }, + { + "epoch": 0.3845971000713097, + "grad_norm": 1.4016800968486762, + "learning_rate": 7.051253808826169e-06, + "loss": 0.5927, + "step": 1618 + }, + { + "epoch": 0.3848347991442833, + "grad_norm": 1.2737052290176687, + "learning_rate": 7.047742114171553e-06, + "loss": 0.6133, + "step": 1619 + }, + { + "epoch": 0.385072498217257, + "grad_norm": 1.2654213841236575, + "learning_rate": 7.0442292054163745e-06, + "loss": 0.635, + "step": 1620 + }, + { + "epoch": 0.3853101972902306, + "grad_norm": 1.1508164771702918, + "learning_rate": 7.040715084643429e-06, + "loss": 0.6454, + "step": 1621 + }, + { + "epoch": 0.3855478963632042, + "grad_norm": 1.4946225993151583, + "learning_rate": 7.037199753936228e-06, + "loss": 0.6391, + "step": 1622 + }, + { + "epoch": 0.3857855954361778, + "grad_norm": 1.6159909455193546, + "learning_rate": 7.033683215379002e-06, + "loss": 0.6281, + "step": 1623 + }, + { + "epoch": 0.3860232945091514, + "grad_norm": 1.4485162960686353, + "learning_rate": 7.030165471056696e-06, + "loss": 0.6133, + "step": 1624 + }, + { + "epoch": 0.386260993582125, + "grad_norm": 1.1528053875730475, + "learning_rate": 7.026646523054972e-06, + "loss": 0.6332, + "step": 1625 + }, + { + "epoch": 0.3864986926550986, + "grad_norm": 1.1500918815365342, + "learning_rate": 7.023126373460202e-06, + "loss": 0.5693, + "step": 1626 + }, + { + "epoch": 0.3867363917280723, + "grad_norm": 1.0246588090577542, + "learning_rate": 7.019605024359475e-06, + "loss": 0.546, + "step": 1627 + }, + { + "epoch": 0.3869740908010459, + "grad_norm": 1.3722949120694947, + "learning_rate": 7.016082477840589e-06, + "loss": 0.5664, + "step": 1628 + }, + { + "epoch": 0.3872117898740195, + "grad_norm": 1.3481838438272098, + "learning_rate": 7.012558735992051e-06, + "loss": 0.6401, + "step": 1629 + }, + { + "epoch": 0.3874494889469931, + "grad_norm": 2.5421372834444624, + "learning_rate": 7.009033800903078e-06, + "loss": 0.6883, + "step": 1630 + }, + { + "epoch": 0.3876871880199667, + "grad_norm": 1.3489381466960801, + "learning_rate": 7.005507674663594e-06, + "loss": 0.6323, + "step": 1631 + }, + { + "epoch": 0.3879248870929403, + "grad_norm": 1.197797035039987, + "learning_rate": 7.00198035936423e-06, + "loss": 0.6227, + "step": 1632 + }, + { + "epoch": 0.3881625861659139, + "grad_norm": 2.3997766034829766, + "learning_rate": 6.998451857096321e-06, + "loss": 0.6001, + "step": 1633 + }, + { + "epoch": 0.3884002852388876, + "grad_norm": 1.1803828612590903, + "learning_rate": 6.994922169951906e-06, + "loss": 0.5893, + "step": 1634 + }, + { + "epoch": 0.3886379843118612, + "grad_norm": 1.5365966829167077, + "learning_rate": 6.991391300023725e-06, + "loss": 0.6235, + "step": 1635 + }, + { + "epoch": 0.3888756833848348, + "grad_norm": 1.0218288185218642, + "learning_rate": 6.987859249405224e-06, + "loss": 0.5555, + "step": 1636 + }, + { + "epoch": 0.3891133824578084, + "grad_norm": 1.0599875026145058, + "learning_rate": 6.984326020190544e-06, + "loss": 0.6274, + "step": 1637 + }, + { + "epoch": 0.389351081530782, + "grad_norm": 1.2717255496299358, + "learning_rate": 6.980791614474527e-06, + "loss": 0.6249, + "step": 1638 + }, + { + "epoch": 0.3895887806037556, + "grad_norm": 1.147504538508881, + "learning_rate": 6.977256034352713e-06, + "loss": 0.5968, + "step": 1639 + }, + { + "epoch": 0.3898264796767293, + "grad_norm": 1.0903976045097978, + "learning_rate": 6.973719281921336e-06, + "loss": 0.6092, + "step": 1640 + }, + { + "epoch": 0.3900641787497029, + "grad_norm": 1.3575612612821835, + "learning_rate": 6.970181359277328e-06, + "loss": 0.5692, + "step": 1641 + }, + { + "epoch": 0.3903018778226765, + "grad_norm": 1.6558969180485714, + "learning_rate": 6.966642268518313e-06, + "loss": 0.5816, + "step": 1642 + }, + { + "epoch": 0.3905395768956501, + "grad_norm": 1.4455213068718662, + "learning_rate": 6.963102011742608e-06, + "loss": 0.6064, + "step": 1643 + }, + { + "epoch": 0.3907772759686237, + "grad_norm": 1.3491011726237432, + "learning_rate": 6.959560591049221e-06, + "loss": 0.5624, + "step": 1644 + }, + { + "epoch": 0.3910149750415973, + "grad_norm": 1.3245013497236533, + "learning_rate": 6.956018008537852e-06, + "loss": 0.5749, + "step": 1645 + }, + { + "epoch": 0.39125267411457093, + "grad_norm": 1.1442536957513634, + "learning_rate": 6.952474266308884e-06, + "loss": 0.6248, + "step": 1646 + }, + { + "epoch": 0.3914903731875446, + "grad_norm": 1.65458423894467, + "learning_rate": 6.948929366463397e-06, + "loss": 0.6318, + "step": 1647 + }, + { + "epoch": 0.3917280722605182, + "grad_norm": 1.3220864781334598, + "learning_rate": 6.945383311103149e-06, + "loss": 0.5688, + "step": 1648 + }, + { + "epoch": 0.3919657713334918, + "grad_norm": 1.6150678731609467, + "learning_rate": 6.941836102330588e-06, + "loss": 0.6048, + "step": 1649 + }, + { + "epoch": 0.3922034704064654, + "grad_norm": 1.1288153451995777, + "learning_rate": 6.938287742248843e-06, + "loss": 0.5958, + "step": 1650 + }, + { + "epoch": 0.392441169479439, + "grad_norm": 1.2391163394053597, + "learning_rate": 6.934738232961728e-06, + "loss": 0.5695, + "step": 1651 + }, + { + "epoch": 0.39267886855241263, + "grad_norm": 1.2991487067562788, + "learning_rate": 6.931187576573733e-06, + "loss": 0.6139, + "step": 1652 + }, + { + "epoch": 0.39291656762538624, + "grad_norm": 1.2901672873182553, + "learning_rate": 6.927635775190038e-06, + "loss": 0.619, + "step": 1653 + }, + { + "epoch": 0.3931542666983599, + "grad_norm": 1.4122434147122394, + "learning_rate": 6.924082830916493e-06, + "loss": 0.5822, + "step": 1654 + }, + { + "epoch": 0.3933919657713335, + "grad_norm": 1.1711425293486566, + "learning_rate": 6.9205287458596305e-06, + "loss": 0.6199, + "step": 1655 + }, + { + "epoch": 0.3936296648443071, + "grad_norm": 1.4086703977384352, + "learning_rate": 6.916973522126655e-06, + "loss": 0.5934, + "step": 1656 + }, + { + "epoch": 0.3938673639172807, + "grad_norm": 1.2067318911446308, + "learning_rate": 6.913417161825449e-06, + "loss": 0.6461, + "step": 1657 + }, + { + "epoch": 0.39410506299025433, + "grad_norm": 1.3165803743349827, + "learning_rate": 6.909859667064571e-06, + "loss": 0.5736, + "step": 1658 + }, + { + "epoch": 0.39434276206322794, + "grad_norm": 1.324929848601222, + "learning_rate": 6.906301039953248e-06, + "loss": 0.6122, + "step": 1659 + }, + { + "epoch": 0.39458046113620154, + "grad_norm": 1.3444692613028137, + "learning_rate": 6.90274128260138e-06, + "loss": 0.6768, + "step": 1660 + }, + { + "epoch": 0.3948181602091752, + "grad_norm": 1.4769509530395613, + "learning_rate": 6.899180397119537e-06, + "loss": 0.6295, + "step": 1661 + }, + { + "epoch": 0.3950558592821488, + "grad_norm": 1.1089129089367786, + "learning_rate": 6.89561838561896e-06, + "loss": 0.6398, + "step": 1662 + }, + { + "epoch": 0.3952935583551224, + "grad_norm": 1.1705625116334932, + "learning_rate": 6.892055250211552e-06, + "loss": 0.5677, + "step": 1663 + }, + { + "epoch": 0.39553125742809603, + "grad_norm": 1.2104748359164996, + "learning_rate": 6.88849099300989e-06, + "loss": 0.5682, + "step": 1664 + }, + { + "epoch": 0.39576895650106964, + "grad_norm": 1.2258257931807746, + "learning_rate": 6.884925616127208e-06, + "loss": 0.6154, + "step": 1665 + }, + { + "epoch": 0.39600665557404324, + "grad_norm": 1.2838260452944084, + "learning_rate": 6.881359121677412e-06, + "loss": 0.6273, + "step": 1666 + }, + { + "epoch": 0.39624435464701685, + "grad_norm": 1.9817733180308856, + "learning_rate": 6.877791511775064e-06, + "loss": 0.5434, + "step": 1667 + }, + { + "epoch": 0.3964820537199905, + "grad_norm": 1.1415847140727422, + "learning_rate": 6.87422278853539e-06, + "loss": 0.6085, + "step": 1668 + }, + { + "epoch": 0.3967197527929641, + "grad_norm": 1.1807235062561758, + "learning_rate": 6.8706529540742775e-06, + "loss": 0.5816, + "step": 1669 + }, + { + "epoch": 0.3969574518659377, + "grad_norm": 1.3363667106488133, + "learning_rate": 6.86708201050827e-06, + "loss": 0.6215, + "step": 1670 + }, + { + "epoch": 0.39719515093891133, + "grad_norm": 1.0729404611876183, + "learning_rate": 6.86350995995457e-06, + "loss": 0.55, + "step": 1671 + }, + { + "epoch": 0.39743285001188494, + "grad_norm": 1.2379682759869208, + "learning_rate": 6.859936804531039e-06, + "loss": 0.6369, + "step": 1672 + }, + { + "epoch": 0.39767054908485855, + "grad_norm": 1.3034702939999545, + "learning_rate": 6.856362546356185e-06, + "loss": 0.6237, + "step": 1673 + }, + { + "epoch": 0.3979082481578322, + "grad_norm": 1.5467208926512261, + "learning_rate": 6.852787187549182e-06, + "loss": 0.6534, + "step": 1674 + }, + { + "epoch": 0.3981459472308058, + "grad_norm": 1.2817004629591544, + "learning_rate": 6.849210730229846e-06, + "loss": 0.6299, + "step": 1675 + }, + { + "epoch": 0.3983836463037794, + "grad_norm": 1.269643173625089, + "learning_rate": 6.845633176518651e-06, + "loss": 0.6537, + "step": 1676 + }, + { + "epoch": 0.39862134537675303, + "grad_norm": 1.1801324709157486, + "learning_rate": 6.842054528536717e-06, + "loss": 0.6276, + "step": 1677 + }, + { + "epoch": 0.39885904444972664, + "grad_norm": 1.1593585234763004, + "learning_rate": 6.838474788405815e-06, + "loss": 0.6087, + "step": 1678 + }, + { + "epoch": 0.39909674352270025, + "grad_norm": 1.535491192383633, + "learning_rate": 6.834893958248361e-06, + "loss": 0.6142, + "step": 1679 + }, + { + "epoch": 0.39933444259567386, + "grad_norm": 1.3589686625357638, + "learning_rate": 6.831312040187424e-06, + "loss": 0.6723, + "step": 1680 + }, + { + "epoch": 0.3995721416686475, + "grad_norm": 1.4030127243639279, + "learning_rate": 6.827729036346706e-06, + "loss": 0.6211, + "step": 1681 + }, + { + "epoch": 0.3998098407416211, + "grad_norm": 1.5026326172124482, + "learning_rate": 6.824144948850567e-06, + "loss": 0.6427, + "step": 1682 + }, + { + "epoch": 0.40004753981459473, + "grad_norm": 1.1149420831958232, + "learning_rate": 6.820559779823999e-06, + "loss": 0.5674, + "step": 1683 + }, + { + "epoch": 0.40028523888756834, + "grad_norm": 1.3463611663055761, + "learning_rate": 6.816973531392639e-06, + "loss": 0.6739, + "step": 1684 + }, + { + "epoch": 0.40052293796054195, + "grad_norm": 1.117795841757387, + "learning_rate": 6.813386205682764e-06, + "loss": 0.6533, + "step": 1685 + }, + { + "epoch": 0.40076063703351555, + "grad_norm": 1.3930558274186762, + "learning_rate": 6.809797804821289e-06, + "loss": 0.5587, + "step": 1686 + }, + { + "epoch": 0.40099833610648916, + "grad_norm": 1.182606007003811, + "learning_rate": 6.806208330935766e-06, + "loss": 0.5351, + "step": 1687 + }, + { + "epoch": 0.4012360351794628, + "grad_norm": 1.1056122303331817, + "learning_rate": 6.802617786154386e-06, + "loss": 0.6005, + "step": 1688 + }, + { + "epoch": 0.40147373425243643, + "grad_norm": 1.2020735867515189, + "learning_rate": 6.79902617260597e-06, + "loss": 0.6119, + "step": 1689 + }, + { + "epoch": 0.40171143332541004, + "grad_norm": 1.3677956730251446, + "learning_rate": 6.795433492419979e-06, + "loss": 0.6478, + "step": 1690 + }, + { + "epoch": 0.40194913239838365, + "grad_norm": 1.6997188536409416, + "learning_rate": 6.7918397477265e-06, + "loss": 0.6091, + "step": 1691 + }, + { + "epoch": 0.40218683147135725, + "grad_norm": 1.2597617625156339, + "learning_rate": 6.788244940656259e-06, + "loss": 0.5915, + "step": 1692 + }, + { + "epoch": 0.40242453054433086, + "grad_norm": 1.1901979414097654, + "learning_rate": 6.784649073340601e-06, + "loss": 0.6078, + "step": 1693 + }, + { + "epoch": 0.40266222961730447, + "grad_norm": 1.3108483274309837, + "learning_rate": 6.78105214791151e-06, + "loss": 0.6925, + "step": 1694 + }, + { + "epoch": 0.40289992869027813, + "grad_norm": 1.4008049253634771, + "learning_rate": 6.77745416650159e-06, + "loss": 0.6254, + "step": 1695 + }, + { + "epoch": 0.40313762776325174, + "grad_norm": 1.0990109710876281, + "learning_rate": 6.773855131244077e-06, + "loss": 0.6103, + "step": 1696 + }, + { + "epoch": 0.40337532683622535, + "grad_norm": 1.3656747034692749, + "learning_rate": 6.770255044272826e-06, + "loss": 0.6193, + "step": 1697 + }, + { + "epoch": 0.40361302590919895, + "grad_norm": 1.1094896788289592, + "learning_rate": 6.766653907722321e-06, + "loss": 0.5823, + "step": 1698 + }, + { + "epoch": 0.40385072498217256, + "grad_norm": 1.2298213060409184, + "learning_rate": 6.763051723727663e-06, + "loss": 0.6238, + "step": 1699 + }, + { + "epoch": 0.40408842405514617, + "grad_norm": 1.378547340840267, + "learning_rate": 6.7594484944245795e-06, + "loss": 0.556, + "step": 1700 + }, + { + "epoch": 0.40408842405514617, + "eval_loss": 0.6133027672767639, + "eval_runtime": 284.8205, + "eval_samples_per_second": 38.192, + "eval_steps_per_second": 1.194, + "step": 1700 + }, + { + "epoch": 0.4043261231281198, + "grad_norm": 1.1886354840850426, + "learning_rate": 6.755844221949414e-06, + "loss": 0.6296, + "step": 1701 + }, + { + "epoch": 0.40456382220109344, + "grad_norm": 1.3163982820362556, + "learning_rate": 6.75223890843913e-06, + "loss": 0.6008, + "step": 1702 + }, + { + "epoch": 0.40480152127406704, + "grad_norm": 1.16923969837126, + "learning_rate": 6.748632556031306e-06, + "loss": 0.6654, + "step": 1703 + }, + { + "epoch": 0.40503922034704065, + "grad_norm": 1.3714200317832606, + "learning_rate": 6.745025166864141e-06, + "loss": 0.606, + "step": 1704 + }, + { + "epoch": 0.40527691942001426, + "grad_norm": 1.261352215487846, + "learning_rate": 6.741416743076443e-06, + "loss": 0.6225, + "step": 1705 + }, + { + "epoch": 0.40551461849298787, + "grad_norm": 1.1473061964862477, + "learning_rate": 6.73780728680764e-06, + "loss": 0.5926, + "step": 1706 + }, + { + "epoch": 0.4057523175659615, + "grad_norm": 1.4348214629324194, + "learning_rate": 6.734196800197763e-06, + "loss": 0.6135, + "step": 1707 + }, + { + "epoch": 0.4059900166389351, + "grad_norm": 1.233353980357576, + "learning_rate": 6.730585285387465e-06, + "loss": 0.6112, + "step": 1708 + }, + { + "epoch": 0.40622771571190874, + "grad_norm": 1.1944881127230755, + "learning_rate": 6.726972744518001e-06, + "loss": 0.5681, + "step": 1709 + }, + { + "epoch": 0.40646541478488235, + "grad_norm": 1.174793725003285, + "learning_rate": 6.723359179731236e-06, + "loss": 0.6354, + "step": 1710 + }, + { + "epoch": 0.40670311385785596, + "grad_norm": 1.156425379631541, + "learning_rate": 6.719744593169642e-06, + "loss": 0.5594, + "step": 1711 + }, + { + "epoch": 0.40694081293082957, + "grad_norm": 1.564455847413731, + "learning_rate": 6.716128986976297e-06, + "loss": 0.546, + "step": 1712 + }, + { + "epoch": 0.4071785120038032, + "grad_norm": 1.133066966307272, + "learning_rate": 6.712512363294885e-06, + "loss": 0.5579, + "step": 1713 + }, + { + "epoch": 0.4074162110767768, + "grad_norm": 1.3966273138847454, + "learning_rate": 6.708894724269692e-06, + "loss": 0.6546, + "step": 1714 + }, + { + "epoch": 0.40765391014975044, + "grad_norm": 1.2275374831037429, + "learning_rate": 6.7052760720456055e-06, + "loss": 0.6104, + "step": 1715 + }, + { + "epoch": 0.40789160922272405, + "grad_norm": 1.1870815195928257, + "learning_rate": 6.701656408768116e-06, + "loss": 0.6461, + "step": 1716 + }, + { + "epoch": 0.40812930829569766, + "grad_norm": 1.2950753442058982, + "learning_rate": 6.698035736583307e-06, + "loss": 0.6765, + "step": 1717 + }, + { + "epoch": 0.40836700736867126, + "grad_norm": 1.4545400516918272, + "learning_rate": 6.69441405763787e-06, + "loss": 0.6523, + "step": 1718 + }, + { + "epoch": 0.40860470644164487, + "grad_norm": 1.3332976726266232, + "learning_rate": 6.690791374079086e-06, + "loss": 0.6128, + "step": 1719 + }, + { + "epoch": 0.4088424055146185, + "grad_norm": 1.1695506020601563, + "learning_rate": 6.687167688054834e-06, + "loss": 0.5824, + "step": 1720 + }, + { + "epoch": 0.4090801045875921, + "grad_norm": 1.317115801820895, + "learning_rate": 6.683543001713588e-06, + "loss": 0.6283, + "step": 1721 + }, + { + "epoch": 0.40931780366056575, + "grad_norm": 1.1484442024580124, + "learning_rate": 6.679917317204414e-06, + "loss": 0.6348, + "step": 1722 + }, + { + "epoch": 0.40955550273353936, + "grad_norm": 1.2496816437159666, + "learning_rate": 6.67629063667697e-06, + "loss": 0.5722, + "step": 1723 + }, + { + "epoch": 0.40979320180651296, + "grad_norm": 1.2802909044704327, + "learning_rate": 6.672662962281504e-06, + "loss": 0.5801, + "step": 1724 + }, + { + "epoch": 0.41003090087948657, + "grad_norm": 1.2522436081295003, + "learning_rate": 6.669034296168855e-06, + "loss": 0.6475, + "step": 1725 + }, + { + "epoch": 0.4102685999524602, + "grad_norm": 1.2251499514563609, + "learning_rate": 6.6654046404904495e-06, + "loss": 0.6031, + "step": 1726 + }, + { + "epoch": 0.4105062990254338, + "grad_norm": 1.3870789459650759, + "learning_rate": 6.6617739973982985e-06, + "loss": 0.5654, + "step": 1727 + }, + { + "epoch": 0.4107439980984074, + "grad_norm": 1.243050744072572, + "learning_rate": 6.658142369045002e-06, + "loss": 0.5363, + "step": 1728 + }, + { + "epoch": 0.41098169717138106, + "grad_norm": 1.1930828770584496, + "learning_rate": 6.6545097575837405e-06, + "loss": 0.6267, + "step": 1729 + }, + { + "epoch": 0.41121939624435466, + "grad_norm": 1.1516724551086497, + "learning_rate": 6.650876165168281e-06, + "loss": 0.6112, + "step": 1730 + }, + { + "epoch": 0.41145709531732827, + "grad_norm": 1.2690590951269962, + "learning_rate": 6.647241593952969e-06, + "loss": 0.6232, + "step": 1731 + }, + { + "epoch": 0.4116947943903019, + "grad_norm": 1.6246910886780845, + "learning_rate": 6.643606046092732e-06, + "loss": 0.6001, + "step": 1732 + }, + { + "epoch": 0.4119324934632755, + "grad_norm": 1.0616726652666646, + "learning_rate": 6.639969523743075e-06, + "loss": 0.5853, + "step": 1733 + }, + { + "epoch": 0.4121701925362491, + "grad_norm": 1.2806527781521033, + "learning_rate": 6.636332029060084e-06, + "loss": 0.5845, + "step": 1734 + }, + { + "epoch": 0.4124078916092227, + "grad_norm": 1.2060707691384536, + "learning_rate": 6.6326935642004165e-06, + "loss": 0.5627, + "step": 1735 + }, + { + "epoch": 0.41264559068219636, + "grad_norm": 1.1490804873403224, + "learning_rate": 6.629054131321312e-06, + "loss": 0.6315, + "step": 1736 + }, + { + "epoch": 0.41288328975516997, + "grad_norm": 1.098785024440525, + "learning_rate": 6.625413732580577e-06, + "loss": 0.5703, + "step": 1737 + }, + { + "epoch": 0.4131209888281436, + "grad_norm": 1.2453480218976, + "learning_rate": 6.621772370136595e-06, + "loss": 0.6492, + "step": 1738 + }, + { + "epoch": 0.4133586879011172, + "grad_norm": 1.4132635941899607, + "learning_rate": 6.618130046148319e-06, + "loss": 0.6387, + "step": 1739 + }, + { + "epoch": 0.4135963869740908, + "grad_norm": 1.2981728678796498, + "learning_rate": 6.614486762775274e-06, + "loss": 0.631, + "step": 1740 + }, + { + "epoch": 0.4138340860470644, + "grad_norm": 1.5891947348458828, + "learning_rate": 6.610842522177549e-06, + "loss": 0.5982, + "step": 1741 + }, + { + "epoch": 0.414071785120038, + "grad_norm": 1.325036774817121, + "learning_rate": 6.607197326515808e-06, + "loss": 0.5932, + "step": 1742 + }, + { + "epoch": 0.41430948419301167, + "grad_norm": 1.1116816751522447, + "learning_rate": 6.6035511779512764e-06, + "loss": 0.5998, + "step": 1743 + }, + { + "epoch": 0.4145471832659853, + "grad_norm": 1.2395854465328835, + "learning_rate": 6.599904078645745e-06, + "loss": 0.5441, + "step": 1744 + }, + { + "epoch": 0.4147848823389589, + "grad_norm": 1.4163979149448744, + "learning_rate": 6.596256030761567e-06, + "loss": 0.6231, + "step": 1745 + }, + { + "epoch": 0.4150225814119325, + "grad_norm": 1.3171114506594737, + "learning_rate": 6.592607036461663e-06, + "loss": 0.5616, + "step": 1746 + }, + { + "epoch": 0.4152602804849061, + "grad_norm": 2.200753805306882, + "learning_rate": 6.588957097909509e-06, + "loss": 0.6429, + "step": 1747 + }, + { + "epoch": 0.4154979795578797, + "grad_norm": 1.3939113042807212, + "learning_rate": 6.585306217269145e-06, + "loss": 0.5649, + "step": 1748 + }, + { + "epoch": 0.41573567863085337, + "grad_norm": 1.233318097225646, + "learning_rate": 6.581654396705166e-06, + "loss": 0.5594, + "step": 1749 + }, + { + "epoch": 0.415973377703827, + "grad_norm": 1.2102569762756403, + "learning_rate": 6.578001638382728e-06, + "loss": 0.5846, + "step": 1750 + }, + { + "epoch": 0.4162110767768006, + "grad_norm": 1.2160667230879676, + "learning_rate": 6.57434794446754e-06, + "loss": 0.6264, + "step": 1751 + }, + { + "epoch": 0.4164487758497742, + "grad_norm": 1.3309303554281997, + "learning_rate": 6.570693317125868e-06, + "loss": 0.608, + "step": 1752 + }, + { + "epoch": 0.4166864749227478, + "grad_norm": 1.2165685302552989, + "learning_rate": 6.567037758524529e-06, + "loss": 0.6036, + "step": 1753 + }, + { + "epoch": 0.4169241739957214, + "grad_norm": 1.2464240711886363, + "learning_rate": 6.563381270830894e-06, + "loss": 0.6017, + "step": 1754 + }, + { + "epoch": 0.417161873068695, + "grad_norm": 1.2082109673363173, + "learning_rate": 6.559723856212885e-06, + "loss": 0.5962, + "step": 1755 + }, + { + "epoch": 0.4173995721416687, + "grad_norm": 1.472925107697074, + "learning_rate": 6.556065516838972e-06, + "loss": 0.5643, + "step": 1756 + }, + { + "epoch": 0.4176372712146423, + "grad_norm": 1.2446920315425474, + "learning_rate": 6.552406254878175e-06, + "loss": 0.5636, + "step": 1757 + }, + { + "epoch": 0.4178749702876159, + "grad_norm": 1.3061378488552113, + "learning_rate": 6.548746072500059e-06, + "loss": 0.6226, + "step": 1758 + }, + { + "epoch": 0.4181126693605895, + "grad_norm": 1.3021553805925503, + "learning_rate": 6.545084971874738e-06, + "loss": 0.6132, + "step": 1759 + }, + { + "epoch": 0.4183503684335631, + "grad_norm": 1.251603688356654, + "learning_rate": 6.541422955172865e-06, + "loss": 0.6329, + "step": 1760 + }, + { + "epoch": 0.4185880675065367, + "grad_norm": 1.147313464631716, + "learning_rate": 6.537760024565643e-06, + "loss": 0.6509, + "step": 1761 + }, + { + "epoch": 0.4188257665795103, + "grad_norm": 1.161100012466609, + "learning_rate": 6.534096182224809e-06, + "loss": 0.6225, + "step": 1762 + }, + { + "epoch": 0.419063465652484, + "grad_norm": 1.5058151662123938, + "learning_rate": 6.53043143032265e-06, + "loss": 0.6508, + "step": 1763 + }, + { + "epoch": 0.4193011647254576, + "grad_norm": 1.5539211144153706, + "learning_rate": 6.5267657710319835e-06, + "loss": 0.6151, + "step": 1764 + }, + { + "epoch": 0.4195388637984312, + "grad_norm": 1.291935143986499, + "learning_rate": 6.5230992065261685e-06, + "loss": 0.5803, + "step": 1765 + }, + { + "epoch": 0.4197765628714048, + "grad_norm": 1.0875362032823737, + "learning_rate": 6.519431738979103e-06, + "loss": 0.5773, + "step": 1766 + }, + { + "epoch": 0.4200142619443784, + "grad_norm": 1.5200341386255265, + "learning_rate": 6.515763370565218e-06, + "loss": 0.6378, + "step": 1767 + }, + { + "epoch": 0.420251961017352, + "grad_norm": 1.1898798679354203, + "learning_rate": 6.5120941034594774e-06, + "loss": 0.6411, + "step": 1768 + }, + { + "epoch": 0.4204896600903256, + "grad_norm": 1.295906193493902, + "learning_rate": 6.508423939837381e-06, + "loss": 0.5972, + "step": 1769 + }, + { + "epoch": 0.4207273591632993, + "grad_norm": 1.1672889806698217, + "learning_rate": 6.504752881874955e-06, + "loss": 0.6014, + "step": 1770 + }, + { + "epoch": 0.4209650582362729, + "grad_norm": 1.2928934803432222, + "learning_rate": 6.501080931748764e-06, + "loss": 0.6021, + "step": 1771 + }, + { + "epoch": 0.4212027573092465, + "grad_norm": 1.1303715473169968, + "learning_rate": 6.4974080916358935e-06, + "loss": 0.5614, + "step": 1772 + }, + { + "epoch": 0.4214404563822201, + "grad_norm": 1.6051022883958528, + "learning_rate": 6.493734363713964e-06, + "loss": 0.6837, + "step": 1773 + }, + { + "epoch": 0.4216781554551937, + "grad_norm": 1.2676824167099554, + "learning_rate": 6.490059750161114e-06, + "loss": 0.6111, + "step": 1774 + }, + { + "epoch": 0.4219158545281673, + "grad_norm": 1.2204987465759312, + "learning_rate": 6.486384253156014e-06, + "loss": 0.592, + "step": 1775 + }, + { + "epoch": 0.42215355360114093, + "grad_norm": 1.2186542294157243, + "learning_rate": 6.482707874877855e-06, + "loss": 0.6451, + "step": 1776 + }, + { + "epoch": 0.4223912526741146, + "grad_norm": 1.3997941320829792, + "learning_rate": 6.4790306175063535e-06, + "loss": 0.6117, + "step": 1777 + }, + { + "epoch": 0.4226289517470882, + "grad_norm": 1.2643277528343335, + "learning_rate": 6.475352483221742e-06, + "loss": 0.6031, + "step": 1778 + }, + { + "epoch": 0.4228666508200618, + "grad_norm": 0.9802448882105571, + "learning_rate": 6.471673474204777e-06, + "loss": 0.5885, + "step": 1779 + }, + { + "epoch": 0.4231043498930354, + "grad_norm": 1.2977581935927913, + "learning_rate": 6.4679935926367335e-06, + "loss": 0.6243, + "step": 1780 + }, + { + "epoch": 0.423342048966009, + "grad_norm": 1.2581299301017672, + "learning_rate": 6.464312840699403e-06, + "loss": 0.5997, + "step": 1781 + }, + { + "epoch": 0.42357974803898263, + "grad_norm": 1.2074801344004524, + "learning_rate": 6.460631220575092e-06, + "loss": 0.5418, + "step": 1782 + }, + { + "epoch": 0.42381744711195624, + "grad_norm": 1.2201326989872303, + "learning_rate": 6.456948734446624e-06, + "loss": 0.5871, + "step": 1783 + }, + { + "epoch": 0.4240551461849299, + "grad_norm": 1.1906356149427682, + "learning_rate": 6.453265384497334e-06, + "loss": 0.6158, + "step": 1784 + }, + { + "epoch": 0.4242928452579035, + "grad_norm": 1.1929544640927696, + "learning_rate": 6.449581172911069e-06, + "loss": 0.6437, + "step": 1785 + }, + { + "epoch": 0.4245305443308771, + "grad_norm": 1.2877172018404999, + "learning_rate": 6.445896101872189e-06, + "loss": 0.6045, + "step": 1786 + }, + { + "epoch": 0.4247682434038507, + "grad_norm": 1.1430509165722447, + "learning_rate": 6.442210173565562e-06, + "loss": 0.5918, + "step": 1787 + }, + { + "epoch": 0.4250059424768243, + "grad_norm": 1.2694186987214204, + "learning_rate": 6.438523390176565e-06, + "loss": 0.5857, + "step": 1788 + }, + { + "epoch": 0.42524364154979793, + "grad_norm": 1.4106948690550583, + "learning_rate": 6.43483575389108e-06, + "loss": 0.6321, + "step": 1789 + }, + { + "epoch": 0.4254813406227716, + "grad_norm": 1.7028026324434216, + "learning_rate": 6.431147266895496e-06, + "loss": 0.6216, + "step": 1790 + }, + { + "epoch": 0.4257190396957452, + "grad_norm": 1.1916991688455243, + "learning_rate": 6.427457931376712e-06, + "loss": 0.5308, + "step": 1791 + }, + { + "epoch": 0.4259567387687188, + "grad_norm": 1.3093795212524022, + "learning_rate": 6.423767749522116e-06, + "loss": 0.6597, + "step": 1792 + }, + { + "epoch": 0.4261944378416924, + "grad_norm": 1.2078656827804846, + "learning_rate": 6.420076723519615e-06, + "loss": 0.572, + "step": 1793 + }, + { + "epoch": 0.426432136914666, + "grad_norm": 1.288941690618705, + "learning_rate": 6.4163848555576e-06, + "loss": 0.6161, + "step": 1794 + }, + { + "epoch": 0.42666983598763963, + "grad_norm": 1.5358476927241855, + "learning_rate": 6.412692147824976e-06, + "loss": 0.6209, + "step": 1795 + }, + { + "epoch": 0.42690753506061324, + "grad_norm": 1.2786331390680228, + "learning_rate": 6.408998602511135e-06, + "loss": 0.5888, + "step": 1796 + }, + { + "epoch": 0.4271452341335869, + "grad_norm": 1.1982565712713893, + "learning_rate": 6.405304221805972e-06, + "loss": 0.5719, + "step": 1797 + }, + { + "epoch": 0.4273829332065605, + "grad_norm": 1.247992840629287, + "learning_rate": 6.401609007899872e-06, + "loss": 0.6569, + "step": 1798 + }, + { + "epoch": 0.4276206322795341, + "grad_norm": 1.5213363580682053, + "learning_rate": 6.39791296298372e-06, + "loss": 0.641, + "step": 1799 + }, + { + "epoch": 0.4278583313525077, + "grad_norm": 1.1067167154702613, + "learning_rate": 6.3942160892488896e-06, + "loss": 0.5604, + "step": 1800 + }, + { + "epoch": 0.4278583313525077, + "eval_loss": 0.609208881855011, + "eval_runtime": 285.2575, + "eval_samples_per_second": 38.134, + "eval_steps_per_second": 1.192, + "step": 1800 + }, + { + "epoch": 0.42809603042548133, + "grad_norm": 1.3540239195321746, + "learning_rate": 6.390518388887246e-06, + "loss": 0.6807, + "step": 1801 + }, + { + "epoch": 0.42833372949845494, + "grad_norm": 1.2103835430934147, + "learning_rate": 6.386819864091146e-06, + "loss": 0.5678, + "step": 1802 + }, + { + "epoch": 0.42857142857142855, + "grad_norm": 1.2056683561464911, + "learning_rate": 6.3831205170534365e-06, + "loss": 0.5837, + "step": 1803 + }, + { + "epoch": 0.4288091276444022, + "grad_norm": 1.1694843098124694, + "learning_rate": 6.379420349967447e-06, + "loss": 0.6033, + "step": 1804 + }, + { + "epoch": 0.4290468267173758, + "grad_norm": 1.1392727037577606, + "learning_rate": 6.375719365026996e-06, + "loss": 0.6218, + "step": 1805 + }, + { + "epoch": 0.4292845257903494, + "grad_norm": 1.338795695397152, + "learning_rate": 6.37201756442639e-06, + "loss": 0.5908, + "step": 1806 + }, + { + "epoch": 0.42952222486332303, + "grad_norm": 1.2477432381383138, + "learning_rate": 6.368314950360416e-06, + "loss": 0.5875, + "step": 1807 + }, + { + "epoch": 0.42975992393629664, + "grad_norm": 1.2808210499890698, + "learning_rate": 6.36461152502434e-06, + "loss": 0.5832, + "step": 1808 + }, + { + "epoch": 0.42999762300927025, + "grad_norm": 1.6456818863612062, + "learning_rate": 6.3609072906139154e-06, + "loss": 0.5893, + "step": 1809 + }, + { + "epoch": 0.43023532208224385, + "grad_norm": 1.253395690643272, + "learning_rate": 6.3572022493253715e-06, + "loss": 0.6227, + "step": 1810 + }, + { + "epoch": 0.4304730211552175, + "grad_norm": 1.4777298040285491, + "learning_rate": 6.353496403355419e-06, + "loss": 0.6021, + "step": 1811 + }, + { + "epoch": 0.4307107202281911, + "grad_norm": 1.241749225526803, + "learning_rate": 6.3497897549012385e-06, + "loss": 0.5821, + "step": 1812 + }, + { + "epoch": 0.43094841930116473, + "grad_norm": 1.3709449933079154, + "learning_rate": 6.3460823061604984e-06, + "loss": 0.6003, + "step": 1813 + }, + { + "epoch": 0.43118611837413834, + "grad_norm": 1.2889469316622355, + "learning_rate": 6.342374059331328e-06, + "loss": 0.5816, + "step": 1814 + }, + { + "epoch": 0.43142381744711195, + "grad_norm": 1.348366759932187, + "learning_rate": 6.3386650166123406e-06, + "loss": 0.6482, + "step": 1815 + }, + { + "epoch": 0.43166151652008555, + "grad_norm": 1.126231603961987, + "learning_rate": 6.334955180202615e-06, + "loss": 0.5581, + "step": 1816 + }, + { + "epoch": 0.43189921559305916, + "grad_norm": 1.2715545513097517, + "learning_rate": 6.331244552301705e-06, + "loss": 0.6489, + "step": 1817 + }, + { + "epoch": 0.4321369146660328, + "grad_norm": 1.1809216133454918, + "learning_rate": 6.32753313510963e-06, + "loss": 0.6179, + "step": 1818 + }, + { + "epoch": 0.43237461373900643, + "grad_norm": 1.1043391329238679, + "learning_rate": 6.323820930826879e-06, + "loss": 0.5942, + "step": 1819 + }, + { + "epoch": 0.43261231281198004, + "grad_norm": 1.3428776123061188, + "learning_rate": 6.32010794165441e-06, + "loss": 0.5847, + "step": 1820 + }, + { + "epoch": 0.43285001188495364, + "grad_norm": 1.1685014782596184, + "learning_rate": 6.316394169793638e-06, + "loss": 0.5683, + "step": 1821 + }, + { + "epoch": 0.43308771095792725, + "grad_norm": 1.2564399463691638, + "learning_rate": 6.312679617446453e-06, + "loss": 0.6047, + "step": 1822 + }, + { + "epoch": 0.43332541003090086, + "grad_norm": 1.3416229149743928, + "learning_rate": 6.308964286815203e-06, + "loss": 0.5973, + "step": 1823 + }, + { + "epoch": 0.4335631091038745, + "grad_norm": 1.581516649950066, + "learning_rate": 6.3052481801026935e-06, + "loss": 0.5784, + "step": 1824 + }, + { + "epoch": 0.43380080817684813, + "grad_norm": 1.3948715720174483, + "learning_rate": 6.301531299512195e-06, + "loss": 0.6296, + "step": 1825 + }, + { + "epoch": 0.43403850724982174, + "grad_norm": 1.4189962582351543, + "learning_rate": 6.297813647247437e-06, + "loss": 0.5569, + "step": 1826 + }, + { + "epoch": 0.43427620632279534, + "grad_norm": 1.256369266666289, + "learning_rate": 6.294095225512604e-06, + "loss": 0.6451, + "step": 1827 + }, + { + "epoch": 0.43451390539576895, + "grad_norm": 1.1866545856703032, + "learning_rate": 6.290376036512339e-06, + "loss": 0.5938, + "step": 1828 + }, + { + "epoch": 0.43475160446874256, + "grad_norm": 1.229467761249601, + "learning_rate": 6.286656082451736e-06, + "loss": 0.5875, + "step": 1829 + }, + { + "epoch": 0.43498930354171617, + "grad_norm": 1.123089186568768, + "learning_rate": 6.2829353655363505e-06, + "loss": 0.5914, + "step": 1830 + }, + { + "epoch": 0.43522700261468983, + "grad_norm": 1.107668538548992, + "learning_rate": 6.279213887972179e-06, + "loss": 0.5725, + "step": 1831 + }, + { + "epoch": 0.43546470168766344, + "grad_norm": 1.3194717813102, + "learning_rate": 6.275491651965678e-06, + "loss": 0.5906, + "step": 1832 + }, + { + "epoch": 0.43570240076063704, + "grad_norm": 1.2308833496500207, + "learning_rate": 6.271768659723751e-06, + "loss": 0.5566, + "step": 1833 + }, + { + "epoch": 0.43594009983361065, + "grad_norm": 1.326606843398089, + "learning_rate": 6.26804491345375e-06, + "loss": 0.573, + "step": 1834 + }, + { + "epoch": 0.43617779890658426, + "grad_norm": 1.4102716893704248, + "learning_rate": 6.264320415363473e-06, + "loss": 0.6829, + "step": 1835 + }, + { + "epoch": 0.43641549797955786, + "grad_norm": 1.3128976431130452, + "learning_rate": 6.260595167661166e-06, + "loss": 0.6121, + "step": 1836 + }, + { + "epoch": 0.43665319705253147, + "grad_norm": 1.2312792951051132, + "learning_rate": 6.2568691725555144e-06, + "loss": 0.5938, + "step": 1837 + }, + { + "epoch": 0.43689089612550513, + "grad_norm": 1.4100858374422707, + "learning_rate": 6.253142432255653e-06, + "loss": 0.6512, + "step": 1838 + }, + { + "epoch": 0.43712859519847874, + "grad_norm": 1.1177363199042611, + "learning_rate": 6.249414948971154e-06, + "loss": 0.6016, + "step": 1839 + }, + { + "epoch": 0.43736629427145235, + "grad_norm": 1.0946832764425742, + "learning_rate": 6.245686724912036e-06, + "loss": 0.5686, + "step": 1840 + }, + { + "epoch": 0.43760399334442596, + "grad_norm": 1.2189374819788859, + "learning_rate": 6.241957762288746e-06, + "loss": 0.578, + "step": 1841 + }, + { + "epoch": 0.43784169241739956, + "grad_norm": 1.1711918905606216, + "learning_rate": 6.238228063312181e-06, + "loss": 0.5706, + "step": 1842 + }, + { + "epoch": 0.43807939149037317, + "grad_norm": 1.3585824080557205, + "learning_rate": 6.234497630193666e-06, + "loss": 0.646, + "step": 1843 + }, + { + "epoch": 0.4383170905633468, + "grad_norm": 1.1993811009998978, + "learning_rate": 6.230766465144966e-06, + "loss": 0.5883, + "step": 1844 + }, + { + "epoch": 0.43855478963632044, + "grad_norm": 1.1858495203001518, + "learning_rate": 6.227034570378278e-06, + "loss": 0.6228, + "step": 1845 + }, + { + "epoch": 0.43879248870929405, + "grad_norm": 1.2370009993585573, + "learning_rate": 6.223301948106231e-06, + "loss": 0.5724, + "step": 1846 + }, + { + "epoch": 0.43903018778226766, + "grad_norm": 1.1468102148796255, + "learning_rate": 6.219568600541886e-06, + "loss": 0.5953, + "step": 1847 + }, + { + "epoch": 0.43926788685524126, + "grad_norm": 1.126638725971421, + "learning_rate": 6.215834529898737e-06, + "loss": 0.5939, + "step": 1848 + }, + { + "epoch": 0.43950558592821487, + "grad_norm": 1.0969373788576617, + "learning_rate": 6.2120997383907015e-06, + "loss": 0.5742, + "step": 1849 + }, + { + "epoch": 0.4397432850011885, + "grad_norm": 1.5780407148907132, + "learning_rate": 6.208364228232128e-06, + "loss": 0.6061, + "step": 1850 + }, + { + "epoch": 0.4399809840741621, + "grad_norm": 1.5237788178953877, + "learning_rate": 6.204628001637788e-06, + "loss": 0.5981, + "step": 1851 + }, + { + "epoch": 0.44021868314713575, + "grad_norm": 1.3367546722628276, + "learning_rate": 6.200891060822884e-06, + "loss": 0.5742, + "step": 1852 + }, + { + "epoch": 0.44045638222010935, + "grad_norm": 1.2945539911597614, + "learning_rate": 6.1971534080030325e-06, + "loss": 0.5827, + "step": 1853 + }, + { + "epoch": 0.44069408129308296, + "grad_norm": 1.4212551758032406, + "learning_rate": 6.193415045394282e-06, + "loss": 0.5919, + "step": 1854 + }, + { + "epoch": 0.44093178036605657, + "grad_norm": 1.2286824798979321, + "learning_rate": 6.189675975213094e-06, + "loss": 0.6053, + "step": 1855 + }, + { + "epoch": 0.4411694794390302, + "grad_norm": 1.3216871596237507, + "learning_rate": 6.185936199676356e-06, + "loss": 0.6105, + "step": 1856 + }, + { + "epoch": 0.4414071785120038, + "grad_norm": 1.1994428953730265, + "learning_rate": 6.182195721001366e-06, + "loss": 0.588, + "step": 1857 + }, + { + "epoch": 0.4416448775849774, + "grad_norm": 1.2711951260494903, + "learning_rate": 6.178454541405849e-06, + "loss": 0.6472, + "step": 1858 + }, + { + "epoch": 0.44188257665795105, + "grad_norm": 1.2046689173138043, + "learning_rate": 6.174712663107933e-06, + "loss": 0.5724, + "step": 1859 + }, + { + "epoch": 0.44212027573092466, + "grad_norm": 1.3496518596358063, + "learning_rate": 6.1709700883261735e-06, + "loss": 0.6818, + "step": 1860 + }, + { + "epoch": 0.44235797480389827, + "grad_norm": 1.3823700903153897, + "learning_rate": 6.1672268192795285e-06, + "loss": 0.637, + "step": 1861 + }, + { + "epoch": 0.4425956738768719, + "grad_norm": 1.3017691843118135, + "learning_rate": 6.163482858187373e-06, + "loss": 0.6139, + "step": 1862 + }, + { + "epoch": 0.4428333729498455, + "grad_norm": 1.185846828280475, + "learning_rate": 6.159738207269491e-06, + "loss": 0.5687, + "step": 1863 + }, + { + "epoch": 0.4430710720228191, + "grad_norm": 1.1985200931360394, + "learning_rate": 6.155992868746078e-06, + "loss": 0.6336, + "step": 1864 + }, + { + "epoch": 0.44330877109579275, + "grad_norm": 1.5171239691816927, + "learning_rate": 6.15224684483773e-06, + "loss": 0.5906, + "step": 1865 + }, + { + "epoch": 0.44354647016876636, + "grad_norm": 1.070992400035297, + "learning_rate": 6.148500137765457e-06, + "loss": 0.5649, + "step": 1866 + }, + { + "epoch": 0.44378416924173997, + "grad_norm": 1.256522627315609, + "learning_rate": 6.144752749750671e-06, + "loss": 0.6068, + "step": 1867 + }, + { + "epoch": 0.4440218683147136, + "grad_norm": 1.347228313354527, + "learning_rate": 6.1410046830151885e-06, + "loss": 0.6084, + "step": 1868 + }, + { + "epoch": 0.4442595673876872, + "grad_norm": 1.1796473006149557, + "learning_rate": 6.1372559397812244e-06, + "loss": 0.6068, + "step": 1869 + }, + { + "epoch": 0.4444972664606608, + "grad_norm": 1.3892004172581018, + "learning_rate": 6.133506522271403e-06, + "loss": 0.599, + "step": 1870 + }, + { + "epoch": 0.4447349655336344, + "grad_norm": 1.1349275405842125, + "learning_rate": 6.129756432708739e-06, + "loss": 0.5966, + "step": 1871 + }, + { + "epoch": 0.44497266460660806, + "grad_norm": 1.0618701816448604, + "learning_rate": 6.126005673316653e-06, + "loss": 0.6371, + "step": 1872 + }, + { + "epoch": 0.44521036367958167, + "grad_norm": 1.1694365006882326, + "learning_rate": 6.122254246318957e-06, + "loss": 0.5976, + "step": 1873 + }, + { + "epoch": 0.4454480627525553, + "grad_norm": 1.210968725162738, + "learning_rate": 6.118502153939866e-06, + "loss": 0.6131, + "step": 1874 + }, + { + "epoch": 0.4456857618255289, + "grad_norm": 1.6645822393203848, + "learning_rate": 6.11474939840398e-06, + "loss": 0.6077, + "step": 1875 + }, + { + "epoch": 0.4459234608985025, + "grad_norm": 1.3346316681456594, + "learning_rate": 6.110995981936299e-06, + "loss": 0.6321, + "step": 1876 + }, + { + "epoch": 0.4461611599714761, + "grad_norm": 4.6210214352258685, + "learning_rate": 6.107241906762214e-06, + "loss": 0.62, + "step": 1877 + }, + { + "epoch": 0.4463988590444497, + "grad_norm": 1.1874602875730624, + "learning_rate": 6.103487175107508e-06, + "loss": 0.6066, + "step": 1878 + }, + { + "epoch": 0.44663655811742337, + "grad_norm": 1.238058036988086, + "learning_rate": 6.099731789198344e-06, + "loss": 0.62, + "step": 1879 + }, + { + "epoch": 0.446874257190397, + "grad_norm": 1.3505778985615207, + "learning_rate": 6.095975751261286e-06, + "loss": 0.5902, + "step": 1880 + }, + { + "epoch": 0.4471119562633706, + "grad_norm": 1.2339693031166519, + "learning_rate": 6.092219063523275e-06, + "loss": 0.5772, + "step": 1881 + }, + { + "epoch": 0.4473496553363442, + "grad_norm": 2.913097863803471, + "learning_rate": 6.088461728211642e-06, + "loss": 0.5969, + "step": 1882 + }, + { + "epoch": 0.4475873544093178, + "grad_norm": 1.2185833776726684, + "learning_rate": 6.0847037475541015e-06, + "loss": 0.6063, + "step": 1883 + }, + { + "epoch": 0.4478250534822914, + "grad_norm": 1.1641362855912283, + "learning_rate": 6.0809451237787485e-06, + "loss": 0.6179, + "step": 1884 + }, + { + "epoch": 0.448062752555265, + "grad_norm": 1.2703246501997127, + "learning_rate": 6.077185859114059e-06, + "loss": 0.618, + "step": 1885 + }, + { + "epoch": 0.44830045162823867, + "grad_norm": 1.2826544368629718, + "learning_rate": 6.073425955788894e-06, + "loss": 0.5452, + "step": 1886 + }, + { + "epoch": 0.4485381507012123, + "grad_norm": 1.1869037357415715, + "learning_rate": 6.0696654160324875e-06, + "loss": 0.538, + "step": 1887 + }, + { + "epoch": 0.4487758497741859, + "grad_norm": 1.2447574395122942, + "learning_rate": 6.065904242074453e-06, + "loss": 0.5776, + "step": 1888 + }, + { + "epoch": 0.4490135488471595, + "grad_norm": 1.364273566665758, + "learning_rate": 6.06214243614478e-06, + "loss": 0.6158, + "step": 1889 + }, + { + "epoch": 0.4492512479201331, + "grad_norm": 1.2044766015082, + "learning_rate": 6.058380000473833e-06, + "loss": 0.5829, + "step": 1890 + }, + { + "epoch": 0.4494889469931067, + "grad_norm": 1.3003044250700955, + "learning_rate": 6.05461693729235e-06, + "loss": 0.61, + "step": 1891 + }, + { + "epoch": 0.4497266460660803, + "grad_norm": 1.4096684563118127, + "learning_rate": 6.05085324883144e-06, + "loss": 0.5697, + "step": 1892 + }, + { + "epoch": 0.449964345139054, + "grad_norm": 1.3871680154718717, + "learning_rate": 6.047088937322582e-06, + "loss": 0.6299, + "step": 1893 + }, + { + "epoch": 0.4502020442120276, + "grad_norm": 1.5524238982398995, + "learning_rate": 6.043324004997629e-06, + "loss": 0.5751, + "step": 1894 + }, + { + "epoch": 0.4504397432850012, + "grad_norm": 1.3522398410560852, + "learning_rate": 6.039558454088796e-06, + "loss": 0.5976, + "step": 1895 + }, + { + "epoch": 0.4506774423579748, + "grad_norm": 1.4693292685016404, + "learning_rate": 6.03579228682867e-06, + "loss": 0.6011, + "step": 1896 + }, + { + "epoch": 0.4509151414309484, + "grad_norm": 1.107303657433106, + "learning_rate": 6.0320255054501985e-06, + "loss": 0.5873, + "step": 1897 + }, + { + "epoch": 0.451152840503922, + "grad_norm": 1.2807765529822432, + "learning_rate": 6.028258112186698e-06, + "loss": 0.5512, + "step": 1898 + }, + { + "epoch": 0.4513905395768957, + "grad_norm": 1.316281760141796, + "learning_rate": 6.024490109271842e-06, + "loss": 0.6, + "step": 1899 + }, + { + "epoch": 0.4516282386498693, + "grad_norm": 1.175653209268442, + "learning_rate": 6.0207214989396734e-06, + "loss": 0.5701, + "step": 1900 + }, + { + "epoch": 0.4516282386498693, + "eval_loss": 0.6042894124984741, + "eval_runtime": 290.5014, + "eval_samples_per_second": 37.446, + "eval_steps_per_second": 1.17, + "step": 1900 + }, + { + "epoch": 0.4518659377228429, + "grad_norm": 1.2210042353043198, + "learning_rate": 6.016952283424586e-06, + "loss": 0.5276, + "step": 1901 + }, + { + "epoch": 0.4521036367958165, + "grad_norm": 1.4180682843551689, + "learning_rate": 6.013182464961341e-06, + "loss": 0.6316, + "step": 1902 + }, + { + "epoch": 0.4523413358687901, + "grad_norm": 1.1985974372474362, + "learning_rate": 6.009412045785051e-06, + "loss": 0.5705, + "step": 1903 + }, + { + "epoch": 0.4525790349417637, + "grad_norm": 1.1905962057429254, + "learning_rate": 6.005641028131188e-06, + "loss": 0.6161, + "step": 1904 + }, + { + "epoch": 0.4528167340147373, + "grad_norm": 1.5918145465932496, + "learning_rate": 6.001869414235577e-06, + "loss": 0.583, + "step": 1905 + }, + { + "epoch": 0.453054433087711, + "grad_norm": 1.163327994928937, + "learning_rate": 5.998097206334399e-06, + "loss": 0.5724, + "step": 1906 + }, + { + "epoch": 0.4532921321606846, + "grad_norm": 1.1773253754635729, + "learning_rate": 5.994324406664184e-06, + "loss": 0.6522, + "step": 1907 + }, + { + "epoch": 0.4535298312336582, + "grad_norm": 1.1524776169805813, + "learning_rate": 5.990551017461815e-06, + "loss": 0.5748, + "step": 1908 + }, + { + "epoch": 0.4537675303066318, + "grad_norm": 1.304471230002249, + "learning_rate": 5.986777040964521e-06, + "loss": 0.6351, + "step": 1909 + }, + { + "epoch": 0.4540052293796054, + "grad_norm": 1.2400533060780092, + "learning_rate": 5.983002479409886e-06, + "loss": 0.6035, + "step": 1910 + }, + { + "epoch": 0.454242928452579, + "grad_norm": 1.2375793230718664, + "learning_rate": 5.9792273350358354e-06, + "loss": 0.624, + "step": 1911 + }, + { + "epoch": 0.4544806275255526, + "grad_norm": 1.5054232871942337, + "learning_rate": 5.975451610080643e-06, + "loss": 0.5837, + "step": 1912 + }, + { + "epoch": 0.4547183265985263, + "grad_norm": 1.3453400392458765, + "learning_rate": 5.971675306782923e-06, + "loss": 0.5997, + "step": 1913 + }, + { + "epoch": 0.4549560256714999, + "grad_norm": 1.494454407923926, + "learning_rate": 5.967898427381636e-06, + "loss": 0.5926, + "step": 1914 + }, + { + "epoch": 0.4551937247444735, + "grad_norm": 1.1909092010550726, + "learning_rate": 5.964120974116085e-06, + "loss": 0.5902, + "step": 1915 + }, + { + "epoch": 0.4554314238174471, + "grad_norm": 1.1392817428834006, + "learning_rate": 5.960342949225909e-06, + "loss": 0.5746, + "step": 1916 + }, + { + "epoch": 0.4556691228904207, + "grad_norm": 1.0954705395567428, + "learning_rate": 5.956564354951091e-06, + "loss": 0.5598, + "step": 1917 + }, + { + "epoch": 0.4559068219633943, + "grad_norm": 1.246924584721644, + "learning_rate": 5.952785193531946e-06, + "loss": 0.6365, + "step": 1918 + }, + { + "epoch": 0.45614452103636793, + "grad_norm": 1.2893419697076727, + "learning_rate": 5.9490054672091305e-06, + "loss": 0.592, + "step": 1919 + }, + { + "epoch": 0.4563822201093416, + "grad_norm": 1.131882790116975, + "learning_rate": 5.945225178223631e-06, + "loss": 0.5657, + "step": 1920 + }, + { + "epoch": 0.4566199191823152, + "grad_norm": 1.1343252584942005, + "learning_rate": 5.941444328816775e-06, + "loss": 0.6213, + "step": 1921 + }, + { + "epoch": 0.4568576182552888, + "grad_norm": 1.5092574110379713, + "learning_rate": 5.937662921230212e-06, + "loss": 0.5773, + "step": 1922 + }, + { + "epoch": 0.4570953173282624, + "grad_norm": 1.3042838372255399, + "learning_rate": 5.933880957705932e-06, + "loss": 0.6171, + "step": 1923 + }, + { + "epoch": 0.457333016401236, + "grad_norm": 1.2495697591538433, + "learning_rate": 5.930098440486248e-06, + "loss": 0.5904, + "step": 1924 + }, + { + "epoch": 0.45757071547420963, + "grad_norm": 1.3770347134153498, + "learning_rate": 5.926315371813807e-06, + "loss": 0.6373, + "step": 1925 + }, + { + "epoch": 0.45780841454718324, + "grad_norm": 1.2802236649629086, + "learning_rate": 5.922531753931575e-06, + "loss": 0.6118, + "step": 1926 + }, + { + "epoch": 0.4580461136201569, + "grad_norm": 1.4151344594018538, + "learning_rate": 5.918747589082853e-06, + "loss": 0.5706, + "step": 1927 + }, + { + "epoch": 0.4582838126931305, + "grad_norm": 1.2598695559732982, + "learning_rate": 5.914962879511259e-06, + "loss": 0.5968, + "step": 1928 + }, + { + "epoch": 0.4585215117661041, + "grad_norm": 1.1161868750568913, + "learning_rate": 5.911177627460739e-06, + "loss": 0.57, + "step": 1929 + }, + { + "epoch": 0.4587592108390777, + "grad_norm": 1.1695155921632332, + "learning_rate": 5.907391835175556e-06, + "loss": 0.5608, + "step": 1930 + }, + { + "epoch": 0.45899690991205133, + "grad_norm": 1.3055255614623045, + "learning_rate": 5.903605504900296e-06, + "loss": 0.5997, + "step": 1931 + }, + { + "epoch": 0.45923460898502494, + "grad_norm": 1.356659606925978, + "learning_rate": 5.899818638879865e-06, + "loss": 0.5822, + "step": 1932 + }, + { + "epoch": 0.45947230805799855, + "grad_norm": 1.644607927733652, + "learning_rate": 5.896031239359485e-06, + "loss": 0.6385, + "step": 1933 + }, + { + "epoch": 0.4597100071309722, + "grad_norm": 1.2635360060440095, + "learning_rate": 5.892243308584693e-06, + "loss": 0.5847, + "step": 1934 + }, + { + "epoch": 0.4599477062039458, + "grad_norm": 1.1087530558760537, + "learning_rate": 5.888454848801345e-06, + "loss": 0.5711, + "step": 1935 + }, + { + "epoch": 0.4601854052769194, + "grad_norm": 1.447875956938966, + "learning_rate": 5.8846658622556055e-06, + "loss": 0.6166, + "step": 1936 + }, + { + "epoch": 0.46042310434989303, + "grad_norm": 1.324482215083244, + "learning_rate": 5.880876351193956e-06, + "loss": 0.5789, + "step": 1937 + }, + { + "epoch": 0.46066080342286664, + "grad_norm": 1.2460990512066645, + "learning_rate": 5.877086317863185e-06, + "loss": 0.6403, + "step": 1938 + }, + { + "epoch": 0.46089850249584025, + "grad_norm": 1.238335026261735, + "learning_rate": 5.8732957645103946e-06, + "loss": 0.6093, + "step": 1939 + }, + { + "epoch": 0.4611362015688139, + "grad_norm": 1.2804461795274884, + "learning_rate": 5.869504693382992e-06, + "loss": 0.6231, + "step": 1940 + }, + { + "epoch": 0.4613739006417875, + "grad_norm": 1.2153603922633087, + "learning_rate": 5.865713106728695e-06, + "loss": 0.5941, + "step": 1941 + }, + { + "epoch": 0.4616115997147611, + "grad_norm": 1.1872205103025297, + "learning_rate": 5.861921006795522e-06, + "loss": 0.5949, + "step": 1942 + }, + { + "epoch": 0.46184929878773473, + "grad_norm": 1.2264257151219657, + "learning_rate": 5.8581283958317995e-06, + "loss": 0.5824, + "step": 1943 + }, + { + "epoch": 0.46208699786070834, + "grad_norm": 1.394990888503453, + "learning_rate": 5.854335276086154e-06, + "loss": 0.61, + "step": 1944 + }, + { + "epoch": 0.46232469693368194, + "grad_norm": 1.1399243267956691, + "learning_rate": 5.85054164980752e-06, + "loss": 0.5427, + "step": 1945 + }, + { + "epoch": 0.46256239600665555, + "grad_norm": 1.2260428134847159, + "learning_rate": 5.846747519245123e-06, + "loss": 0.5839, + "step": 1946 + }, + { + "epoch": 0.4628000950796292, + "grad_norm": 1.265869732725906, + "learning_rate": 5.842952886648496e-06, + "loss": 0.6283, + "step": 1947 + }, + { + "epoch": 0.4630377941526028, + "grad_norm": 1.4222273978861646, + "learning_rate": 5.839157754267464e-06, + "loss": 0.655, + "step": 1948 + }, + { + "epoch": 0.46327549322557643, + "grad_norm": 1.1005462299776745, + "learning_rate": 5.835362124352152e-06, + "loss": 0.5678, + "step": 1949 + }, + { + "epoch": 0.46351319229855004, + "grad_norm": 1.3121163824639899, + "learning_rate": 5.831565999152976e-06, + "loss": 0.6429, + "step": 1950 + }, + { + "epoch": 0.46375089137152364, + "grad_norm": 1.2210284146824035, + "learning_rate": 5.82776938092065e-06, + "loss": 0.5788, + "step": 1951 + }, + { + "epoch": 0.46398859044449725, + "grad_norm": 1.211347817571114, + "learning_rate": 5.823972271906177e-06, + "loss": 0.5881, + "step": 1952 + }, + { + "epoch": 0.46422628951747086, + "grad_norm": 1.315728832213456, + "learning_rate": 5.820174674360855e-06, + "loss": 0.5887, + "step": 1953 + }, + { + "epoch": 0.4644639885904445, + "grad_norm": 1.0739294540386193, + "learning_rate": 5.816376590536265e-06, + "loss": 0.6051, + "step": 1954 + }, + { + "epoch": 0.46470168766341813, + "grad_norm": 1.2659748210013455, + "learning_rate": 5.812578022684282e-06, + "loss": 0.6332, + "step": 1955 + }, + { + "epoch": 0.46493938673639174, + "grad_norm": 1.460850120325146, + "learning_rate": 5.808778973057066e-06, + "loss": 0.5983, + "step": 1956 + }, + { + "epoch": 0.46517708580936534, + "grad_norm": 1.112517543569675, + "learning_rate": 5.804979443907065e-06, + "loss": 0.5739, + "step": 1957 + }, + { + "epoch": 0.46541478488233895, + "grad_norm": 1.1858271561822848, + "learning_rate": 5.801179437487005e-06, + "loss": 0.5441, + "step": 1958 + }, + { + "epoch": 0.46565248395531256, + "grad_norm": 1.1459332249328833, + "learning_rate": 5.797378956049905e-06, + "loss": 0.5932, + "step": 1959 + }, + { + "epoch": 0.46589018302828616, + "grad_norm": 1.1790263369546667, + "learning_rate": 5.793578001849054e-06, + "loss": 0.612, + "step": 1960 + }, + { + "epoch": 0.4661278821012598, + "grad_norm": 1.2512708315401768, + "learning_rate": 5.78977657713803e-06, + "loss": 0.5888, + "step": 1961 + }, + { + "epoch": 0.46636558117423343, + "grad_norm": 1.0982447390382737, + "learning_rate": 5.785974684170685e-06, + "loss": 0.6614, + "step": 1962 + }, + { + "epoch": 0.46660328024720704, + "grad_norm": 1.1605475677170372, + "learning_rate": 5.782172325201155e-06, + "loss": 0.5982, + "step": 1963 + }, + { + "epoch": 0.46684097932018065, + "grad_norm": 1.1073217382565355, + "learning_rate": 5.778369502483844e-06, + "loss": 0.556, + "step": 1964 + }, + { + "epoch": 0.46707867839315426, + "grad_norm": 1.0656680782980998, + "learning_rate": 5.7745662182734364e-06, + "loss": 0.6201, + "step": 1965 + }, + { + "epoch": 0.46731637746612786, + "grad_norm": 1.2056086889301947, + "learning_rate": 5.770762474824887e-06, + "loss": 0.6156, + "step": 1966 + }, + { + "epoch": 0.46755407653910147, + "grad_norm": 1.2999447861087683, + "learning_rate": 5.766958274393428e-06, + "loss": 0.6243, + "step": 1967 + }, + { + "epoch": 0.46779177561207513, + "grad_norm": 1.2874471426859784, + "learning_rate": 5.763153619234556e-06, + "loss": 0.6104, + "step": 1968 + }, + { + "epoch": 0.46802947468504874, + "grad_norm": 1.2014673706912458, + "learning_rate": 5.7593485116040425e-06, + "loss": 0.5998, + "step": 1969 + }, + { + "epoch": 0.46826717375802235, + "grad_norm": 1.2506618534216376, + "learning_rate": 5.755542953757923e-06, + "loss": 0.5294, + "step": 1970 + }, + { + "epoch": 0.46850487283099596, + "grad_norm": 1.1388064548802725, + "learning_rate": 5.7517369479525044e-06, + "loss": 0.5225, + "step": 1971 + }, + { + "epoch": 0.46874257190396956, + "grad_norm": 2.020983552724873, + "learning_rate": 5.747930496444356e-06, + "loss": 0.6486, + "step": 1972 + }, + { + "epoch": 0.46898027097694317, + "grad_norm": 1.1584383100736744, + "learning_rate": 5.744123601490312e-06, + "loss": 0.5983, + "step": 1973 + }, + { + "epoch": 0.46921797004991683, + "grad_norm": 1.2358809665960488, + "learning_rate": 5.74031626534747e-06, + "loss": 0.6147, + "step": 1974 + }, + { + "epoch": 0.46945566912289044, + "grad_norm": 1.2897030217904553, + "learning_rate": 5.736508490273189e-06, + "loss": 0.6104, + "step": 1975 + }, + { + "epoch": 0.46969336819586405, + "grad_norm": 1.3288228783833391, + "learning_rate": 5.7327002785250864e-06, + "loss": 0.5588, + "step": 1976 + }, + { + "epoch": 0.46993106726883765, + "grad_norm": 1.207156487396507, + "learning_rate": 5.728891632361043e-06, + "loss": 0.5843, + "step": 1977 + }, + { + "epoch": 0.47016876634181126, + "grad_norm": 1.2569401982009234, + "learning_rate": 5.725082554039193e-06, + "loss": 0.5949, + "step": 1978 + }, + { + "epoch": 0.47040646541478487, + "grad_norm": 1.3051974882095265, + "learning_rate": 5.72127304581793e-06, + "loss": 0.6107, + "step": 1979 + }, + { + "epoch": 0.4706441644877585, + "grad_norm": 1.231621612747007, + "learning_rate": 5.717463109955896e-06, + "loss": 0.5729, + "step": 1980 + }, + { + "epoch": 0.47088186356073214, + "grad_norm": 1.319441190852223, + "learning_rate": 5.713652748711997e-06, + "loss": 0.5838, + "step": 1981 + }, + { + "epoch": 0.47111956263370575, + "grad_norm": 1.199871060788161, + "learning_rate": 5.709841964345383e-06, + "loss": 0.5684, + "step": 1982 + }, + { + "epoch": 0.47135726170667935, + "grad_norm": 1.1839379883774743, + "learning_rate": 5.706030759115458e-06, + "loss": 0.6, + "step": 1983 + }, + { + "epoch": 0.47159496077965296, + "grad_norm": 1.2760771626698864, + "learning_rate": 5.702219135281872e-06, + "loss": 0.5521, + "step": 1984 + }, + { + "epoch": 0.47183265985262657, + "grad_norm": 1.2979307753472915, + "learning_rate": 5.698407095104531e-06, + "loss": 0.5555, + "step": 1985 + }, + { + "epoch": 0.4720703589256002, + "grad_norm": 1.2142499435994982, + "learning_rate": 5.694594640843579e-06, + "loss": 0.5523, + "step": 1986 + }, + { + "epoch": 0.4723080579985738, + "grad_norm": 1.3313854355841173, + "learning_rate": 5.690781774759412e-06, + "loss": 0.5982, + "step": 1987 + }, + { + "epoch": 0.47254575707154745, + "grad_norm": 1.3310784831288052, + "learning_rate": 5.686968499112665e-06, + "loss": 0.6168, + "step": 1988 + }, + { + "epoch": 0.47278345614452105, + "grad_norm": 1.205662786714757, + "learning_rate": 5.6831548161642215e-06, + "loss": 0.6087, + "step": 1989 + }, + { + "epoch": 0.47302115521749466, + "grad_norm": 1.5324715413306411, + "learning_rate": 5.6793407281752e-06, + "loss": 0.5905, + "step": 1990 + }, + { + "epoch": 0.47325885429046827, + "grad_norm": 1.4178632945433542, + "learning_rate": 5.675526237406965e-06, + "loss": 0.6312, + "step": 1991 + }, + { + "epoch": 0.4734965533634419, + "grad_norm": 1.2989529743784816, + "learning_rate": 5.671711346121117e-06, + "loss": 0.602, + "step": 1992 + }, + { + "epoch": 0.4737342524364155, + "grad_norm": 1.2798908040542334, + "learning_rate": 5.667896056579495e-06, + "loss": 0.6458, + "step": 1993 + }, + { + "epoch": 0.4739719515093891, + "grad_norm": 1.2167364616360377, + "learning_rate": 5.664080371044172e-06, + "loss": 0.6031, + "step": 1994 + }, + { + "epoch": 0.47420965058236275, + "grad_norm": 1.1172748441398237, + "learning_rate": 5.660264291777458e-06, + "loss": 0.6158, + "step": 1995 + }, + { + "epoch": 0.47444734965533636, + "grad_norm": 1.3615902985603021, + "learning_rate": 5.6564478210418946e-06, + "loss": 0.5896, + "step": 1996 + }, + { + "epoch": 0.47468504872830997, + "grad_norm": 1.3198280293261073, + "learning_rate": 5.65263096110026e-06, + "loss": 0.5841, + "step": 1997 + }, + { + "epoch": 0.4749227478012836, + "grad_norm": 1.390337958878552, + "learning_rate": 5.648813714215554e-06, + "loss": 0.5532, + "step": 1998 + }, + { + "epoch": 0.4751604468742572, + "grad_norm": 1.648899664784063, + "learning_rate": 5.644996082651018e-06, + "loss": 0.6309, + "step": 1999 + }, + { + "epoch": 0.4753981459472308, + "grad_norm": 1.1634863381922331, + "learning_rate": 5.64117806867011e-06, + "loss": 0.565, + "step": 2000 + }, + { + "epoch": 0.4753981459472308, + "eval_loss": 0.6003658771514893, + "eval_runtime": 290.9824, + "eval_samples_per_second": 37.384, + "eval_steps_per_second": 1.168, + "step": 2000 + }, + { + "epoch": 0.4756358450202044, + "grad_norm": 1.0736705436757767, + "learning_rate": 5.637359674536522e-06, + "loss": 0.5596, + "step": 2001 + }, + { + "epoch": 0.47587354409317806, + "grad_norm": 1.155382426355328, + "learning_rate": 5.63354090251417e-06, + "loss": 0.5697, + "step": 2002 + }, + { + "epoch": 0.47611124316615167, + "grad_norm": 1.40498236872398, + "learning_rate": 5.629721754867192e-06, + "loss": 0.6235, + "step": 2003 + }, + { + "epoch": 0.4763489422391253, + "grad_norm": 1.366167488796539, + "learning_rate": 5.625902233859949e-06, + "loss": 0.56, + "step": 2004 + }, + { + "epoch": 0.4765866413120989, + "grad_norm": 1.311683079728917, + "learning_rate": 5.622082341757027e-06, + "loss": 0.6151, + "step": 2005 + }, + { + "epoch": 0.4768243403850725, + "grad_norm": 1.445786211407705, + "learning_rate": 5.618262080823227e-06, + "loss": 0.5663, + "step": 2006 + }, + { + "epoch": 0.4770620394580461, + "grad_norm": 1.2316048307570666, + "learning_rate": 5.614441453323571e-06, + "loss": 0.6109, + "step": 2007 + }, + { + "epoch": 0.4772997385310197, + "grad_norm": 1.34534446392238, + "learning_rate": 5.610620461523301e-06, + "loss": 0.5997, + "step": 2008 + }, + { + "epoch": 0.47753743760399336, + "grad_norm": 1.1799259644322218, + "learning_rate": 5.606799107687868e-06, + "loss": 0.5932, + "step": 2009 + }, + { + "epoch": 0.47777513667696697, + "grad_norm": 1.3233446132773397, + "learning_rate": 5.602977394082947e-06, + "loss": 0.6441, + "step": 2010 + }, + { + "epoch": 0.4780128357499406, + "grad_norm": 1.2572355013740915, + "learning_rate": 5.5991553229744166e-06, + "loss": 0.6431, + "step": 2011 + }, + { + "epoch": 0.4782505348229142, + "grad_norm": 1.2739532572224632, + "learning_rate": 5.595332896628375e-06, + "loss": 0.5887, + "step": 2012 + }, + { + "epoch": 0.4784882338958878, + "grad_norm": 9.546236209818924, + "learning_rate": 5.591510117311128e-06, + "loss": 0.5682, + "step": 2013 + }, + { + "epoch": 0.4787259329688614, + "grad_norm": 1.144682403333344, + "learning_rate": 5.587686987289189e-06, + "loss": 0.5878, + "step": 2014 + }, + { + "epoch": 0.47896363204183506, + "grad_norm": 1.1928897824858558, + "learning_rate": 5.583863508829281e-06, + "loss": 0.5647, + "step": 2015 + }, + { + "epoch": 0.47920133111480867, + "grad_norm": 1.4426143661308546, + "learning_rate": 5.580039684198337e-06, + "loss": 0.5733, + "step": 2016 + }, + { + "epoch": 0.4794390301877823, + "grad_norm": 1.1532640862149632, + "learning_rate": 5.576215515663489e-06, + "loss": 0.5962, + "step": 2017 + }, + { + "epoch": 0.4796767292607559, + "grad_norm": 1.2337612317170945, + "learning_rate": 5.572391005492076e-06, + "loss": 0.5835, + "step": 2018 + }, + { + "epoch": 0.4799144283337295, + "grad_norm": 1.2918042464233208, + "learning_rate": 5.5685661559516395e-06, + "loss": 0.5888, + "step": 2019 + }, + { + "epoch": 0.4801521274067031, + "grad_norm": 1.1944016718964907, + "learning_rate": 5.564740969309923e-06, + "loss": 0.5853, + "step": 2020 + }, + { + "epoch": 0.4803898264796767, + "grad_norm": 1.4662082838810722, + "learning_rate": 5.560915447834868e-06, + "loss": 0.5712, + "step": 2021 + }, + { + "epoch": 0.48062752555265037, + "grad_norm": 1.3112088046724697, + "learning_rate": 5.557089593794616e-06, + "loss": 0.5812, + "step": 2022 + }, + { + "epoch": 0.480865224625624, + "grad_norm": 2.839998700583429, + "learning_rate": 5.553263409457504e-06, + "loss": 0.6156, + "step": 2023 + }, + { + "epoch": 0.4811029236985976, + "grad_norm": 1.2537340409683608, + "learning_rate": 5.549436897092068e-06, + "loss": 0.5896, + "step": 2024 + }, + { + "epoch": 0.4813406227715712, + "grad_norm": 1.2831245382042413, + "learning_rate": 5.545610058967034e-06, + "loss": 0.5577, + "step": 2025 + }, + { + "epoch": 0.4815783218445448, + "grad_norm": 1.2593905616057237, + "learning_rate": 5.541782897351327e-06, + "loss": 0.6144, + "step": 2026 + }, + { + "epoch": 0.4818160209175184, + "grad_norm": 1.2130895464472138, + "learning_rate": 5.537955414514058e-06, + "loss": 0.6013, + "step": 2027 + }, + { + "epoch": 0.482053719990492, + "grad_norm": 1.1386147158971887, + "learning_rate": 5.53412761272453e-06, + "loss": 0.5516, + "step": 2028 + }, + { + "epoch": 0.4822914190634657, + "grad_norm": 1.0907788379379701, + "learning_rate": 5.530299494252238e-06, + "loss": 0.5632, + "step": 2029 + }, + { + "epoch": 0.4825291181364393, + "grad_norm": 1.3762412506739299, + "learning_rate": 5.526471061366861e-06, + "loss": 0.6154, + "step": 2030 + }, + { + "epoch": 0.4827668172094129, + "grad_norm": 1.1475918519567474, + "learning_rate": 5.522642316338268e-06, + "loss": 0.5621, + "step": 2031 + }, + { + "epoch": 0.4830045162823865, + "grad_norm": 1.361953909358766, + "learning_rate": 5.51881326143651e-06, + "loss": 0.5789, + "step": 2032 + }, + { + "epoch": 0.4832422153553601, + "grad_norm": 1.344121517293254, + "learning_rate": 5.514983898931823e-06, + "loss": 0.6098, + "step": 2033 + }, + { + "epoch": 0.4834799144283337, + "grad_norm": 1.1038326142099373, + "learning_rate": 5.511154231094625e-06, + "loss": 0.6046, + "step": 2034 + }, + { + "epoch": 0.4837176135013073, + "grad_norm": 1.3183924974815424, + "learning_rate": 5.507324260195516e-06, + "loss": 0.5761, + "step": 2035 + }, + { + "epoch": 0.483955312574281, + "grad_norm": 1.1660395884009673, + "learning_rate": 5.503493988505275e-06, + "loss": 0.6019, + "step": 2036 + }, + { + "epoch": 0.4841930116472546, + "grad_norm": 1.3908947090851258, + "learning_rate": 5.499663418294858e-06, + "loss": 0.5721, + "step": 2037 + }, + { + "epoch": 0.4844307107202282, + "grad_norm": 1.2650068818884472, + "learning_rate": 5.4958325518354015e-06, + "loss": 0.6022, + "step": 2038 + }, + { + "epoch": 0.4846684097932018, + "grad_norm": 1.247765554911343, + "learning_rate": 5.492001391398214e-06, + "loss": 0.6192, + "step": 2039 + }, + { + "epoch": 0.4849061088661754, + "grad_norm": 1.1738132366297394, + "learning_rate": 5.488169939254781e-06, + "loss": 0.5859, + "step": 2040 + }, + { + "epoch": 0.485143807939149, + "grad_norm": 1.3424436983207442, + "learning_rate": 5.484338197676757e-06, + "loss": 0.6115, + "step": 2041 + }, + { + "epoch": 0.4853815070121226, + "grad_norm": 1.3122808075864456, + "learning_rate": 5.4805061689359754e-06, + "loss": 0.5641, + "step": 2042 + }, + { + "epoch": 0.4856192060850963, + "grad_norm": 1.6412394180374805, + "learning_rate": 5.4766738553044306e-06, + "loss": 0.581, + "step": 2043 + }, + { + "epoch": 0.4858569051580699, + "grad_norm": 1.314430142108364, + "learning_rate": 5.472841259054294e-06, + "loss": 0.6298, + "step": 2044 + }, + { + "epoch": 0.4860946042310435, + "grad_norm": 1.257773966964291, + "learning_rate": 5.4690083824579e-06, + "loss": 0.6082, + "step": 2045 + }, + { + "epoch": 0.4863323033040171, + "grad_norm": 1.447863970683147, + "learning_rate": 5.46517522778775e-06, + "loss": 0.6191, + "step": 2046 + }, + { + "epoch": 0.4865700023769907, + "grad_norm": 1.1794141648291783, + "learning_rate": 5.46134179731651e-06, + "loss": 0.5746, + "step": 2047 + }, + { + "epoch": 0.4868077014499643, + "grad_norm": 1.4015510922147811, + "learning_rate": 5.457508093317013e-06, + "loss": 0.5901, + "step": 2048 + }, + { + "epoch": 0.487045400522938, + "grad_norm": 1.0754880268204787, + "learning_rate": 5.453674118062248e-06, + "loss": 0.6292, + "step": 2049 + }, + { + "epoch": 0.4872830995959116, + "grad_norm": 1.3999224431355979, + "learning_rate": 5.449839873825371e-06, + "loss": 0.6574, + "step": 2050 + }, + { + "epoch": 0.4875207986688852, + "grad_norm": 1.1384998340253671, + "learning_rate": 5.446005362879692e-06, + "loss": 0.5616, + "step": 2051 + }, + { + "epoch": 0.4877584977418588, + "grad_norm": 1.2261238318050687, + "learning_rate": 5.442170587498684e-06, + "loss": 0.606, + "step": 2052 + }, + { + "epoch": 0.4879961968148324, + "grad_norm": 1.2410575631122218, + "learning_rate": 5.4383355499559734e-06, + "loss": 0.5887, + "step": 2053 + }, + { + "epoch": 0.488233895887806, + "grad_norm": 1.282964380510751, + "learning_rate": 5.434500252525343e-06, + "loss": 0.6001, + "step": 2054 + }, + { + "epoch": 0.48847159496077963, + "grad_norm": 1.1308636475521976, + "learning_rate": 5.430664697480731e-06, + "loss": 0.5868, + "step": 2055 + }, + { + "epoch": 0.4887092940337533, + "grad_norm": 1.2793770832942, + "learning_rate": 5.426828887096227e-06, + "loss": 0.5912, + "step": 2056 + }, + { + "epoch": 0.4889469931067269, + "grad_norm": 1.1836296623543048, + "learning_rate": 5.4229928236460705e-06, + "loss": 0.5975, + "step": 2057 + }, + { + "epoch": 0.4891846921797005, + "grad_norm": 1.2443664773375671, + "learning_rate": 5.419156509404656e-06, + "loss": 0.6133, + "step": 2058 + }, + { + "epoch": 0.4894223912526741, + "grad_norm": 1.3255788860679758, + "learning_rate": 5.41531994664652e-06, + "loss": 0.5522, + "step": 2059 + }, + { + "epoch": 0.4896600903256477, + "grad_norm": 1.1298171492058104, + "learning_rate": 5.411483137646353e-06, + "loss": 0.5841, + "step": 2060 + }, + { + "epoch": 0.48989778939862133, + "grad_norm": 1.2381829183884718, + "learning_rate": 5.407646084678985e-06, + "loss": 0.5947, + "step": 2061 + }, + { + "epoch": 0.49013548847159494, + "grad_norm": 1.2988525308781378, + "learning_rate": 5.4038087900193985e-06, + "loss": 0.5785, + "step": 2062 + }, + { + "epoch": 0.4903731875445686, + "grad_norm": 1.2733026189696017, + "learning_rate": 5.399971255942708e-06, + "loss": 0.5902, + "step": 2063 + }, + { + "epoch": 0.4906108866175422, + "grad_norm": 1.23777243056835, + "learning_rate": 5.396133484724184e-06, + "loss": 0.5825, + "step": 2064 + }, + { + "epoch": 0.4908485856905158, + "grad_norm": 1.2377139961086379, + "learning_rate": 5.392295478639226e-06, + "loss": 0.5548, + "step": 2065 + }, + { + "epoch": 0.4910862847634894, + "grad_norm": 1.7572218760721963, + "learning_rate": 5.388457239963379e-06, + "loss": 0.5906, + "step": 2066 + }, + { + "epoch": 0.49132398383646303, + "grad_norm": 1.127891975970903, + "learning_rate": 5.3846187709723195e-06, + "loss": 0.5896, + "step": 2067 + }, + { + "epoch": 0.49156168290943664, + "grad_norm": 1.2519869818105638, + "learning_rate": 5.380780073941871e-06, + "loss": 0.5511, + "step": 2068 + }, + { + "epoch": 0.49179938198241024, + "grad_norm": 1.3137983092656023, + "learning_rate": 5.376941151147983e-06, + "loss": 0.6173, + "step": 2069 + }, + { + "epoch": 0.4920370810553839, + "grad_norm": 1.2465338075212964, + "learning_rate": 5.373102004866744e-06, + "loss": 0.6405, + "step": 2070 + }, + { + "epoch": 0.4922747801283575, + "grad_norm": 1.1019621480877486, + "learning_rate": 5.36926263737437e-06, + "loss": 0.5815, + "step": 2071 + }, + { + "epoch": 0.4925124792013311, + "grad_norm": 1.4265385617471071, + "learning_rate": 5.365423050947214e-06, + "loss": 0.5684, + "step": 2072 + }, + { + "epoch": 0.49275017827430473, + "grad_norm": 1.1206033782246974, + "learning_rate": 5.361583247861756e-06, + "loss": 0.5955, + "step": 2073 + }, + { + "epoch": 0.49298787734727834, + "grad_norm": 1.443888020249758, + "learning_rate": 5.357743230394602e-06, + "loss": 0.578, + "step": 2074 + }, + { + "epoch": 0.49322557642025194, + "grad_norm": 1.1157373055434496, + "learning_rate": 5.3539030008224906e-06, + "loss": 0.5971, + "step": 2075 + }, + { + "epoch": 0.49346327549322555, + "grad_norm": 1.173758150750993, + "learning_rate": 5.3500625614222805e-06, + "loss": 0.6141, + "step": 2076 + }, + { + "epoch": 0.4937009745661992, + "grad_norm": 1.2363747448710702, + "learning_rate": 5.346221914470959e-06, + "loss": 0.5734, + "step": 2077 + }, + { + "epoch": 0.4939386736391728, + "grad_norm": 1.3073195037053555, + "learning_rate": 5.342381062245634e-06, + "loss": 0.5751, + "step": 2078 + }, + { + "epoch": 0.49417637271214643, + "grad_norm": 1.7604945737606088, + "learning_rate": 5.338540007023538e-06, + "loss": 0.5971, + "step": 2079 + }, + { + "epoch": 0.49441407178512004, + "grad_norm": 1.33980296140168, + "learning_rate": 5.334698751082019e-06, + "loss": 0.5648, + "step": 2080 + }, + { + "epoch": 0.49465177085809364, + "grad_norm": 1.2893338338948421, + "learning_rate": 5.330857296698551e-06, + "loss": 0.5471, + "step": 2081 + }, + { + "epoch": 0.49488946993106725, + "grad_norm": 1.1572873821811043, + "learning_rate": 5.327015646150716e-06, + "loss": 0.5625, + "step": 2082 + }, + { + "epoch": 0.49512716900404086, + "grad_norm": 1.4676473836380781, + "learning_rate": 5.323173801716222e-06, + "loss": 0.5886, + "step": 2083 + }, + { + "epoch": 0.4953648680770145, + "grad_norm": 1.3316111942184035, + "learning_rate": 5.319331765672886e-06, + "loss": 0.595, + "step": 2084 + }, + { + "epoch": 0.4956025671499881, + "grad_norm": 1.1155277530189285, + "learning_rate": 5.315489540298642e-06, + "loss": 0.5573, + "step": 2085 + }, + { + "epoch": 0.49584026622296173, + "grad_norm": 1.2829744626392592, + "learning_rate": 5.311647127871532e-06, + "loss": 0.6075, + "step": 2086 + }, + { + "epoch": 0.49607796529593534, + "grad_norm": 1.2262710659683211, + "learning_rate": 5.3078045306697154e-06, + "loss": 0.5078, + "step": 2087 + }, + { + "epoch": 0.49631566436890895, + "grad_norm": 1.2033002994465287, + "learning_rate": 5.3039617509714535e-06, + "loss": 0.5967, + "step": 2088 + }, + { + "epoch": 0.49655336344188256, + "grad_norm": 1.2149701598190654, + "learning_rate": 5.300118791055122e-06, + "loss": 0.6133, + "step": 2089 + }, + { + "epoch": 0.4967910625148562, + "grad_norm": 1.431201304903657, + "learning_rate": 5.2962756531991985e-06, + "loss": 0.5692, + "step": 2090 + }, + { + "epoch": 0.4970287615878298, + "grad_norm": 1.1994953202356946, + "learning_rate": 5.292432339682273e-06, + "loss": 0.5844, + "step": 2091 + }, + { + "epoch": 0.49726646066080343, + "grad_norm": 1.360726516373225, + "learning_rate": 5.288588852783031e-06, + "loss": 0.5397, + "step": 2092 + }, + { + "epoch": 0.49750415973377704, + "grad_norm": 1.2946552814888155, + "learning_rate": 5.284745194780267e-06, + "loss": 0.6008, + "step": 2093 + }, + { + "epoch": 0.49774185880675065, + "grad_norm": 1.2704236348289348, + "learning_rate": 5.280901367952876e-06, + "loss": 0.6125, + "step": 2094 + }, + { + "epoch": 0.49797955787972426, + "grad_norm": 1.5654729479376956, + "learning_rate": 5.27705737457985e-06, + "loss": 0.6035, + "step": 2095 + }, + { + "epoch": 0.49821725695269786, + "grad_norm": 1.1488198223522479, + "learning_rate": 5.273213216940284e-06, + "loss": 0.5667, + "step": 2096 + }, + { + "epoch": 0.4984549560256715, + "grad_norm": 1.2868490290521357, + "learning_rate": 5.2693688973133675e-06, + "loss": 0.5521, + "step": 2097 + }, + { + "epoch": 0.49869265509864513, + "grad_norm": 1.1738391615150408, + "learning_rate": 5.265524417978385e-06, + "loss": 0.6063, + "step": 2098 + }, + { + "epoch": 0.49893035417161874, + "grad_norm": 1.2079622484359538, + "learning_rate": 5.2616797812147205e-06, + "loss": 0.5723, + "step": 2099 + }, + { + "epoch": 0.49916805324459235, + "grad_norm": 1.354397923686889, + "learning_rate": 5.257834989301846e-06, + "loss": 0.6045, + "step": 2100 + }, + { + "epoch": 0.49916805324459235, + "eval_loss": 0.5957692861557007, + "eval_runtime": 290.6481, + "eval_samples_per_second": 37.427, + "eval_steps_per_second": 1.17, + "step": 2100 + }, + { + "epoch": 0.49940575231756595, + "grad_norm": 1.5037276220648303, + "learning_rate": 5.253990044519329e-06, + "loss": 0.6336, + "step": 2101 + }, + { + "epoch": 0.49964345139053956, + "grad_norm": 1.2893490971216537, + "learning_rate": 5.250144949146827e-06, + "loss": 0.5609, + "step": 2102 + }, + { + "epoch": 0.49988115046351317, + "grad_norm": 1.4781439220551056, + "learning_rate": 5.246299705464085e-06, + "loss": 0.613, + "step": 2103 + }, + { + "epoch": 0.5001188495364868, + "grad_norm": 1.6187443448422651, + "learning_rate": 5.242454315750938e-06, + "loss": 0.6036, + "step": 2104 + }, + { + "epoch": 0.5003565486094604, + "grad_norm": 1.5579974302517252, + "learning_rate": 5.238608782287307e-06, + "loss": 0.6346, + "step": 2105 + }, + { + "epoch": 0.500594247682434, + "grad_norm": 1.206303876047271, + "learning_rate": 5.234763107353197e-06, + "loss": 0.5822, + "step": 2106 + }, + { + "epoch": 0.5008319467554077, + "grad_norm": 1.2374308297238557, + "learning_rate": 5.230917293228699e-06, + "loss": 0.536, + "step": 2107 + }, + { + "epoch": 0.5010696458283813, + "grad_norm": 1.3593177683700541, + "learning_rate": 5.227071342193985e-06, + "loss": 0.5523, + "step": 2108 + }, + { + "epoch": 0.5013073449013549, + "grad_norm": 1.3012621649107703, + "learning_rate": 5.223225256529309e-06, + "loss": 0.602, + "step": 2109 + }, + { + "epoch": 0.5015450439743285, + "grad_norm": 1.2827508251422342, + "learning_rate": 5.219379038515001e-06, + "loss": 0.6126, + "step": 2110 + }, + { + "epoch": 0.5017827430473021, + "grad_norm": 1.1509859407967271, + "learning_rate": 5.2155326904314795e-06, + "loss": 0.5929, + "step": 2111 + }, + { + "epoch": 0.5020204421202757, + "grad_norm": 2.8836945155480405, + "learning_rate": 5.211686214559227e-06, + "loss": 0.6298, + "step": 2112 + }, + { + "epoch": 0.5022581411932493, + "grad_norm": 1.200921101119019, + "learning_rate": 5.207839613178814e-06, + "loss": 0.5801, + "step": 2113 + }, + { + "epoch": 0.502495840266223, + "grad_norm": 1.261684106112285, + "learning_rate": 5.203992888570873e-06, + "loss": 0.5302, + "step": 2114 + }, + { + "epoch": 0.5027335393391966, + "grad_norm": 1.1583936588181964, + "learning_rate": 5.200146043016123e-06, + "loss": 0.5945, + "step": 2115 + }, + { + "epoch": 0.5029712384121702, + "grad_norm": 1.3382780033381747, + "learning_rate": 5.1962990787953436e-06, + "loss": 0.5908, + "step": 2116 + }, + { + "epoch": 0.5032089374851438, + "grad_norm": 1.2119442660769175, + "learning_rate": 5.192451998189392e-06, + "loss": 0.5574, + "step": 2117 + }, + { + "epoch": 0.5034466365581174, + "grad_norm": 1.0782856705663333, + "learning_rate": 5.188604803479188e-06, + "loss": 0.5591, + "step": 2118 + }, + { + "epoch": 0.503684335631091, + "grad_norm": 3.013214878308211, + "learning_rate": 5.184757496945726e-06, + "loss": 0.6032, + "step": 2119 + }, + { + "epoch": 0.5039220347040646, + "grad_norm": 1.3079737495631873, + "learning_rate": 5.18091008087006e-06, + "loss": 0.5783, + "step": 2120 + }, + { + "epoch": 0.5041597337770383, + "grad_norm": 1.191683844743298, + "learning_rate": 5.177062557533315e-06, + "loss": 0.6248, + "step": 2121 + }, + { + "epoch": 0.5043974328500119, + "grad_norm": 1.3982740765667458, + "learning_rate": 5.173214929216677e-06, + "loss": 0.5863, + "step": 2122 + }, + { + "epoch": 0.5046351319229855, + "grad_norm": 1.2798173828709127, + "learning_rate": 5.169367198201391e-06, + "loss": 0.629, + "step": 2123 + }, + { + "epoch": 0.5048728309959591, + "grad_norm": 1.120917020095986, + "learning_rate": 5.165519366768769e-06, + "loss": 0.5809, + "step": 2124 + }, + { + "epoch": 0.5051105300689327, + "grad_norm": 1.794898762662154, + "learning_rate": 5.161671437200179e-06, + "loss": 0.5871, + "step": 2125 + }, + { + "epoch": 0.5053482291419064, + "grad_norm": 1.1150602960373017, + "learning_rate": 5.157823411777045e-06, + "loss": 0.5541, + "step": 2126 + }, + { + "epoch": 0.5055859282148799, + "grad_norm": 1.5043457495485226, + "learning_rate": 5.153975292780852e-06, + "loss": 0.5847, + "step": 2127 + }, + { + "epoch": 0.5058236272878536, + "grad_norm": 1.3398893237131224, + "learning_rate": 5.1501270824931415e-06, + "loss": 0.5627, + "step": 2128 + }, + { + "epoch": 0.5060613263608272, + "grad_norm": 1.1324751745653425, + "learning_rate": 5.1462787831955045e-06, + "loss": 0.5452, + "step": 2129 + }, + { + "epoch": 0.5062990254338008, + "grad_norm": 1.1952327994111036, + "learning_rate": 5.142430397169585e-06, + "loss": 0.6172, + "step": 2130 + }, + { + "epoch": 0.5065367245067744, + "grad_norm": 1.638768100195031, + "learning_rate": 5.138581926697083e-06, + "loss": 0.6098, + "step": 2131 + }, + { + "epoch": 0.506774423579748, + "grad_norm": 1.4183760303011832, + "learning_rate": 5.134733374059745e-06, + "loss": 0.6357, + "step": 2132 + }, + { + "epoch": 0.5070121226527217, + "grad_norm": 1.3209705940824847, + "learning_rate": 5.130884741539367e-06, + "loss": 0.5768, + "step": 2133 + }, + { + "epoch": 0.5072498217256952, + "grad_norm": 1.165926546374581, + "learning_rate": 5.127036031417792e-06, + "loss": 0.5685, + "step": 2134 + }, + { + "epoch": 0.5074875207986689, + "grad_norm": 1.1357811601307446, + "learning_rate": 5.123187245976912e-06, + "loss": 0.5919, + "step": 2135 + }, + { + "epoch": 0.5077252198716425, + "grad_norm": 1.1578175020834836, + "learning_rate": 5.119338387498656e-06, + "loss": 0.5962, + "step": 2136 + }, + { + "epoch": 0.5079629189446161, + "grad_norm": 1.271250014495172, + "learning_rate": 5.115489458265006e-06, + "loss": 0.5573, + "step": 2137 + }, + { + "epoch": 0.5082006180175898, + "grad_norm": 1.3514105464434254, + "learning_rate": 5.11164046055798e-06, + "loss": 0.5631, + "step": 2138 + }, + { + "epoch": 0.5084383170905633, + "grad_norm": 1.4812394269152644, + "learning_rate": 5.107791396659637e-06, + "loss": 0.6108, + "step": 2139 + }, + { + "epoch": 0.508676016163537, + "grad_norm": 1.3788587064490623, + "learning_rate": 5.103942268852078e-06, + "loss": 0.622, + "step": 2140 + }, + { + "epoch": 0.5089137152365105, + "grad_norm": 1.2932637718566842, + "learning_rate": 5.100093079417439e-06, + "loss": 0.5483, + "step": 2141 + }, + { + "epoch": 0.5091514143094842, + "grad_norm": 1.1731871487869947, + "learning_rate": 5.0962438306378924e-06, + "loss": 0.5597, + "step": 2142 + }, + { + "epoch": 0.5093891133824578, + "grad_norm": 3.725547373658594, + "learning_rate": 5.09239452479565e-06, + "loss": 0.6175, + "step": 2143 + }, + { + "epoch": 0.5096268124554314, + "grad_norm": 1.2014532504614759, + "learning_rate": 5.088545164172951e-06, + "loss": 0.5443, + "step": 2144 + }, + { + "epoch": 0.5098645115284051, + "grad_norm": 1.1713243086391532, + "learning_rate": 5.084695751052075e-06, + "loss": 0.5368, + "step": 2145 + }, + { + "epoch": 0.5101022106013786, + "grad_norm": 1.1840475123122574, + "learning_rate": 5.080846287715323e-06, + "loss": 0.6134, + "step": 2146 + }, + { + "epoch": 0.5103399096743523, + "grad_norm": 1.1882308385504254, + "learning_rate": 5.0769967764450345e-06, + "loss": 0.5739, + "step": 2147 + }, + { + "epoch": 0.5105776087473259, + "grad_norm": 1.9101909921841156, + "learning_rate": 5.073147219523572e-06, + "loss": 0.6166, + "step": 2148 + }, + { + "epoch": 0.5108153078202995, + "grad_norm": 1.3365866916357172, + "learning_rate": 5.0692976192333295e-06, + "loss": 0.5642, + "step": 2149 + }, + { + "epoch": 0.5110530068932732, + "grad_norm": 1.3916237008153858, + "learning_rate": 5.065447977856723e-06, + "loss": 0.5706, + "step": 2150 + }, + { + "epoch": 0.5112907059662467, + "grad_norm": 1.2463393377418603, + "learning_rate": 5.061598297676192e-06, + "loss": 0.6197, + "step": 2151 + }, + { + "epoch": 0.5115284050392204, + "grad_norm": 1.3697345846821756, + "learning_rate": 5.057748580974204e-06, + "loss": 0.5789, + "step": 2152 + }, + { + "epoch": 0.5117661041121939, + "grad_norm": 1.3616713038875963, + "learning_rate": 5.0538988300332446e-06, + "loss": 0.5907, + "step": 2153 + }, + { + "epoch": 0.5120038031851676, + "grad_norm": 1.360674928892087, + "learning_rate": 5.050049047135819e-06, + "loss": 0.4872, + "step": 2154 + }, + { + "epoch": 0.5122415022581412, + "grad_norm": 1.4299248923729901, + "learning_rate": 5.046199234564455e-06, + "loss": 0.564, + "step": 2155 + }, + { + "epoch": 0.5124792013311148, + "grad_norm": 1.1566201593277299, + "learning_rate": 5.042349394601692e-06, + "loss": 0.6118, + "step": 2156 + }, + { + "epoch": 0.5127169004040885, + "grad_norm": 1.1684745781854358, + "learning_rate": 5.038499529530094e-06, + "loss": 0.6447, + "step": 2157 + }, + { + "epoch": 0.512954599477062, + "grad_norm": 1.2292105252144483, + "learning_rate": 5.034649641632231e-06, + "loss": 0.582, + "step": 2158 + }, + { + "epoch": 0.5131922985500357, + "grad_norm": 1.3679457136421993, + "learning_rate": 5.030799733190694e-06, + "loss": 0.5592, + "step": 2159 + }, + { + "epoch": 0.5134299976230092, + "grad_norm": 1.7132219089642808, + "learning_rate": 5.0269498064880804e-06, + "loss": 0.6056, + "step": 2160 + }, + { + "epoch": 0.5136676966959829, + "grad_norm": 1.48535364397568, + "learning_rate": 5.0230998638070024e-06, + "loss": 0.581, + "step": 2161 + }, + { + "epoch": 0.5139053957689566, + "grad_norm": 1.1362907855585251, + "learning_rate": 5.019249907430079e-06, + "loss": 0.5853, + "step": 2162 + }, + { + "epoch": 0.5141430948419301, + "grad_norm": 1.256564917662599, + "learning_rate": 5.015399939639941e-06, + "loss": 0.6271, + "step": 2163 + }, + { + "epoch": 0.5143807939149038, + "grad_norm": 1.4381550837392827, + "learning_rate": 5.011549962719221e-06, + "loss": 0.6724, + "step": 2164 + }, + { + "epoch": 0.5146184929878773, + "grad_norm": 1.2278262907147328, + "learning_rate": 5.0076999789505625e-06, + "loss": 0.558, + "step": 2165 + }, + { + "epoch": 0.514856192060851, + "grad_norm": 1.1963766852574362, + "learning_rate": 5.003849990616605e-06, + "loss": 0.5436, + "step": 2166 + }, + { + "epoch": 0.5150938911338245, + "grad_norm": 1.3380109967133689, + "learning_rate": 5e-06, + "loss": 0.5982, + "step": 2167 + }, + { + "epoch": 0.5153315902067982, + "grad_norm": 1.1459678044769193, + "learning_rate": 4.996150009383397e-06, + "loss": 0.6338, + "step": 2168 + }, + { + "epoch": 0.5155692892797719, + "grad_norm": 1.2244755696283687, + "learning_rate": 4.99230002104944e-06, + "loss": 0.5336, + "step": 2169 + }, + { + "epoch": 0.5158069883527454, + "grad_norm": 1.224715554895932, + "learning_rate": 4.98845003728078e-06, + "loss": 0.5799, + "step": 2170 + }, + { + "epoch": 0.5160446874257191, + "grad_norm": 1.1004822149927425, + "learning_rate": 4.98460006036006e-06, + "loss": 0.5566, + "step": 2171 + }, + { + "epoch": 0.5162823864986926, + "grad_norm": 1.3009409007559514, + "learning_rate": 4.9807500925699215e-06, + "loss": 0.5643, + "step": 2172 + }, + { + "epoch": 0.5165200855716663, + "grad_norm": 1.157993096260443, + "learning_rate": 4.976900136192998e-06, + "loss": 0.554, + "step": 2173 + }, + { + "epoch": 0.5167577846446398, + "grad_norm": 1.1746390878069457, + "learning_rate": 4.973050193511921e-06, + "loss": 0.6244, + "step": 2174 + }, + { + "epoch": 0.5169954837176135, + "grad_norm": 1.5527088244577332, + "learning_rate": 4.9692002668093075e-06, + "loss": 0.5859, + "step": 2175 + }, + { + "epoch": 0.5172331827905872, + "grad_norm": 1.2989425640797363, + "learning_rate": 4.96535035836777e-06, + "loss": 0.6081, + "step": 2176 + }, + { + "epoch": 0.5174708818635607, + "grad_norm": 1.419331425475817, + "learning_rate": 4.961500470469908e-06, + "loss": 0.5955, + "step": 2177 + }, + { + "epoch": 0.5177085809365344, + "grad_norm": 1.242491388630374, + "learning_rate": 4.957650605398309e-06, + "loss": 0.5414, + "step": 2178 + }, + { + "epoch": 0.5179462800095079, + "grad_norm": 1.1828203671442759, + "learning_rate": 4.953800765435547e-06, + "loss": 0.5683, + "step": 2179 + }, + { + "epoch": 0.5181839790824816, + "grad_norm": 1.2437965070348243, + "learning_rate": 4.949950952864183e-06, + "loss": 0.6044, + "step": 2180 + }, + { + "epoch": 0.5184216781554551, + "grad_norm": 1.2082242963647913, + "learning_rate": 4.946101169966756e-06, + "loss": 0.5705, + "step": 2181 + }, + { + "epoch": 0.5186593772284288, + "grad_norm": 2.078333757829309, + "learning_rate": 4.942251419025797e-06, + "loss": 0.5622, + "step": 2182 + }, + { + "epoch": 0.5188970763014025, + "grad_norm": 1.1792884453460113, + "learning_rate": 4.9384017023238085e-06, + "loss": 0.5523, + "step": 2183 + }, + { + "epoch": 0.519134775374376, + "grad_norm": 1.2150991564867328, + "learning_rate": 4.934552022143279e-06, + "loss": 0.6384, + "step": 2184 + }, + { + "epoch": 0.5193724744473497, + "grad_norm": 1.3522211378489664, + "learning_rate": 4.930702380766671e-06, + "loss": 0.5677, + "step": 2185 + }, + { + "epoch": 0.5196101735203232, + "grad_norm": 1.264475260194322, + "learning_rate": 4.9268527804764285e-06, + "loss": 0.599, + "step": 2186 + }, + { + "epoch": 0.5198478725932969, + "grad_norm": 1.2066747054706044, + "learning_rate": 4.923003223554967e-06, + "loss": 0.5503, + "step": 2187 + }, + { + "epoch": 0.5200855716662705, + "grad_norm": 1.2304020152496573, + "learning_rate": 4.9191537122846785e-06, + "loss": 0.5613, + "step": 2188 + }, + { + "epoch": 0.5203232707392441, + "grad_norm": 1.277206603801028, + "learning_rate": 4.915304248947927e-06, + "loss": 0.6253, + "step": 2189 + }, + { + "epoch": 0.5205609698122178, + "grad_norm": 1.290067092254496, + "learning_rate": 4.91145483582705e-06, + "loss": 0.6482, + "step": 2190 + }, + { + "epoch": 0.5207986688851913, + "grad_norm": 1.2105535609668365, + "learning_rate": 4.907605475204352e-06, + "loss": 0.6011, + "step": 2191 + }, + { + "epoch": 0.521036367958165, + "grad_norm": 1.2483144000017452, + "learning_rate": 4.903756169362109e-06, + "loss": 0.6068, + "step": 2192 + }, + { + "epoch": 0.5212740670311385, + "grad_norm": 1.3839774583072713, + "learning_rate": 4.899906920582563e-06, + "loss": 0.5546, + "step": 2193 + }, + { + "epoch": 0.5215117661041122, + "grad_norm": 1.1702118274575213, + "learning_rate": 4.896057731147924e-06, + "loss": 0.6082, + "step": 2194 + }, + { + "epoch": 0.5217494651770858, + "grad_norm": 1.1494461364625042, + "learning_rate": 4.892208603340364e-06, + "loss": 0.5467, + "step": 2195 + }, + { + "epoch": 0.5219871642500594, + "grad_norm": 1.2031090672505351, + "learning_rate": 4.888359539442023e-06, + "loss": 0.5885, + "step": 2196 + }, + { + "epoch": 0.5222248633230331, + "grad_norm": 1.203936363818319, + "learning_rate": 4.8845105417349955e-06, + "loss": 0.5726, + "step": 2197 + }, + { + "epoch": 0.5224625623960066, + "grad_norm": 1.4276077648089245, + "learning_rate": 4.880661612501346e-06, + "loss": 0.5719, + "step": 2198 + }, + { + "epoch": 0.5227002614689803, + "grad_norm": 1.1964322811506596, + "learning_rate": 4.876812754023092e-06, + "loss": 0.5468, + "step": 2199 + }, + { + "epoch": 0.5229379605419539, + "grad_norm": 1.3234673194045878, + "learning_rate": 4.87296396858221e-06, + "loss": 0.5595, + "step": 2200 + }, + { + "epoch": 0.5229379605419539, + "eval_loss": 0.5927928686141968, + "eval_runtime": 291.4268, + "eval_samples_per_second": 37.327, + "eval_steps_per_second": 1.167, + "step": 2200 + }, + { + "epoch": 0.5231756596149275, + "grad_norm": 1.2646504378329395, + "learning_rate": 4.869115258460636e-06, + "loss": 0.5858, + "step": 2201 + }, + { + "epoch": 0.5234133586879011, + "grad_norm": 1.3479648251182712, + "learning_rate": 4.8652666259402584e-06, + "loss": 0.5823, + "step": 2202 + }, + { + "epoch": 0.5236510577608747, + "grad_norm": 1.4184277398928085, + "learning_rate": 4.861418073302919e-06, + "loss": 0.5301, + "step": 2203 + }, + { + "epoch": 0.5238887568338484, + "grad_norm": 1.5659409936450184, + "learning_rate": 4.857569602830418e-06, + "loss": 0.5769, + "step": 2204 + }, + { + "epoch": 0.5241264559068219, + "grad_norm": 1.4327061851174092, + "learning_rate": 4.853721216804499e-06, + "loss": 0.5855, + "step": 2205 + }, + { + "epoch": 0.5243641549797956, + "grad_norm": 1.1415297910815216, + "learning_rate": 4.849872917506862e-06, + "loss": 0.572, + "step": 2206 + }, + { + "epoch": 0.5246018540527692, + "grad_norm": 1.238056615696645, + "learning_rate": 4.846024707219149e-06, + "loss": 0.5588, + "step": 2207 + }, + { + "epoch": 0.5248395531257428, + "grad_norm": 1.1879779551610925, + "learning_rate": 4.842176588222958e-06, + "loss": 0.5799, + "step": 2208 + }, + { + "epoch": 0.5250772521987164, + "grad_norm": 1.1663677075187018, + "learning_rate": 4.838328562799824e-06, + "loss": 0.6096, + "step": 2209 + }, + { + "epoch": 0.52531495127169, + "grad_norm": 1.2829161442137078, + "learning_rate": 4.834480633231234e-06, + "loss": 0.5737, + "step": 2210 + }, + { + "epoch": 0.5255526503446637, + "grad_norm": 1.8462238184523203, + "learning_rate": 4.830632801798611e-06, + "loss": 0.5854, + "step": 2211 + }, + { + "epoch": 0.5257903494176372, + "grad_norm": 1.0529485890410755, + "learning_rate": 4.826785070783326e-06, + "loss": 0.5623, + "step": 2212 + }, + { + "epoch": 0.5260280484906109, + "grad_norm": 1.2082963510472438, + "learning_rate": 4.8229374424666865e-06, + "loss": 0.5954, + "step": 2213 + }, + { + "epoch": 0.5262657475635845, + "grad_norm": 1.127733219574055, + "learning_rate": 4.819089919129942e-06, + "loss": 0.5113, + "step": 2214 + }, + { + "epoch": 0.5265034466365581, + "grad_norm": 1.2952222453867779, + "learning_rate": 4.815242503054277e-06, + "loss": 0.5661, + "step": 2215 + }, + { + "epoch": 0.5267411457095317, + "grad_norm": 1.6387234872865575, + "learning_rate": 4.8113951965208145e-06, + "loss": 0.6072, + "step": 2216 + }, + { + "epoch": 0.5269788447825053, + "grad_norm": 1.3827646072828697, + "learning_rate": 4.807548001810611e-06, + "loss": 0.5884, + "step": 2217 + }, + { + "epoch": 0.527216543855479, + "grad_norm": 1.2025444000833756, + "learning_rate": 4.803700921204659e-06, + "loss": 0.6183, + "step": 2218 + }, + { + "epoch": 0.5274542429284526, + "grad_norm": 1.3588172934151832, + "learning_rate": 4.799853956983879e-06, + "loss": 0.6293, + "step": 2219 + }, + { + "epoch": 0.5276919420014262, + "grad_norm": 1.2703789768682987, + "learning_rate": 4.796007111429129e-06, + "loss": 0.582, + "step": 2220 + }, + { + "epoch": 0.5279296410743998, + "grad_norm": 1.5882345784138097, + "learning_rate": 4.79216038682119e-06, + "loss": 0.5615, + "step": 2221 + }, + { + "epoch": 0.5281673401473734, + "grad_norm": 1.7190709376929096, + "learning_rate": 4.7883137854407755e-06, + "loss": 0.5555, + "step": 2222 + }, + { + "epoch": 0.5284050392203471, + "grad_norm": 1.368498280075712, + "learning_rate": 4.784467309568524e-06, + "loss": 0.5552, + "step": 2223 + }, + { + "epoch": 0.5286427382933206, + "grad_norm": 1.2559712787044413, + "learning_rate": 4.780620961485001e-06, + "loss": 0.5609, + "step": 2224 + }, + { + "epoch": 0.5288804373662943, + "grad_norm": 1.4318185590965902, + "learning_rate": 4.776774743470695e-06, + "loss": 0.6148, + "step": 2225 + }, + { + "epoch": 0.5291181364392679, + "grad_norm": 1.2589866606138285, + "learning_rate": 4.7729286578060185e-06, + "loss": 0.5982, + "step": 2226 + }, + { + "epoch": 0.5293558355122415, + "grad_norm": 1.2580281856496982, + "learning_rate": 4.7690827067713035e-06, + "loss": 0.5559, + "step": 2227 + }, + { + "epoch": 0.5295935345852151, + "grad_norm": 1.28820975683986, + "learning_rate": 4.7652368926468055e-06, + "loss": 0.5399, + "step": 2228 + }, + { + "epoch": 0.5298312336581887, + "grad_norm": 1.231447482194784, + "learning_rate": 4.761391217712696e-06, + "loss": 0.5346, + "step": 2229 + }, + { + "epoch": 0.5300689327311624, + "grad_norm": 1.178819658949713, + "learning_rate": 4.757545684249065e-06, + "loss": 0.6048, + "step": 2230 + }, + { + "epoch": 0.530306631804136, + "grad_norm": 1.182385962437741, + "learning_rate": 4.753700294535916e-06, + "loss": 0.5488, + "step": 2231 + }, + { + "epoch": 0.5305443308771096, + "grad_norm": 1.293857171077926, + "learning_rate": 4.749855050853174e-06, + "loss": 0.5658, + "step": 2232 + }, + { + "epoch": 0.5307820299500832, + "grad_norm": 1.285667272067441, + "learning_rate": 4.746009955480672e-06, + "loss": 0.5611, + "step": 2233 + }, + { + "epoch": 0.5310197290230568, + "grad_norm": 1.3133699757201192, + "learning_rate": 4.742165010698155e-06, + "loss": 0.5718, + "step": 2234 + }, + { + "epoch": 0.5312574280960304, + "grad_norm": 1.2651120903699151, + "learning_rate": 4.738320218785281e-06, + "loss": 0.6108, + "step": 2235 + }, + { + "epoch": 0.531495127169004, + "grad_norm": 1.2074627481906692, + "learning_rate": 4.734475582021615e-06, + "loss": 0.5563, + "step": 2236 + }, + { + "epoch": 0.5317328262419777, + "grad_norm": 1.567067823790544, + "learning_rate": 4.730631102686635e-06, + "loss": 0.6057, + "step": 2237 + }, + { + "epoch": 0.5319705253149513, + "grad_norm": 1.16841069367857, + "learning_rate": 4.726786783059716e-06, + "loss": 0.5737, + "step": 2238 + }, + { + "epoch": 0.5322082243879249, + "grad_norm": 1.301328079273045, + "learning_rate": 4.7229426254201504e-06, + "loss": 0.6009, + "step": 2239 + }, + { + "epoch": 0.5324459234608985, + "grad_norm": 1.2197377046412032, + "learning_rate": 4.719098632047124e-06, + "loss": 0.5581, + "step": 2240 + }, + { + "epoch": 0.5326836225338721, + "grad_norm": 1.2074683595974758, + "learning_rate": 4.715254805219734e-06, + "loss": 0.5886, + "step": 2241 + }, + { + "epoch": 0.5329213216068457, + "grad_norm": 1.2440125540133398, + "learning_rate": 4.711411147216969e-06, + "loss": 0.5711, + "step": 2242 + }, + { + "epoch": 0.5331590206798194, + "grad_norm": 1.2496828463394485, + "learning_rate": 4.7075676603177285e-06, + "loss": 0.6059, + "step": 2243 + }, + { + "epoch": 0.533396719752793, + "grad_norm": 1.1653421578602863, + "learning_rate": 4.7037243468008014e-06, + "loss": 0.5815, + "step": 2244 + }, + { + "epoch": 0.5336344188257666, + "grad_norm": 1.4899325056871076, + "learning_rate": 4.69988120894488e-06, + "loss": 0.5914, + "step": 2245 + }, + { + "epoch": 0.5338721178987402, + "grad_norm": 1.3074381071505956, + "learning_rate": 4.696038249028547e-06, + "loss": 0.5661, + "step": 2246 + }, + { + "epoch": 0.5341098169717138, + "grad_norm": 1.1589040837713258, + "learning_rate": 4.692195469330286e-06, + "loss": 0.5671, + "step": 2247 + }, + { + "epoch": 0.5343475160446874, + "grad_norm": 1.262057729338904, + "learning_rate": 4.688352872128468e-06, + "loss": 0.5462, + "step": 2248 + }, + { + "epoch": 0.534585215117661, + "grad_norm": 1.6331637588020866, + "learning_rate": 4.68451045970136e-06, + "loss": 0.6166, + "step": 2249 + }, + { + "epoch": 0.5348229141906347, + "grad_norm": 1.1538590455240887, + "learning_rate": 4.6806682343271145e-06, + "loss": 0.5834, + "step": 2250 + }, + { + "epoch": 0.5350606132636083, + "grad_norm": 1.251144104897622, + "learning_rate": 4.676826198283779e-06, + "loss": 0.5216, + "step": 2251 + }, + { + "epoch": 0.5352983123365819, + "grad_norm": 1.229603625617296, + "learning_rate": 4.672984353849285e-06, + "loss": 0.6432, + "step": 2252 + }, + { + "epoch": 0.5355360114095555, + "grad_norm": 1.223646440152381, + "learning_rate": 4.669142703301452e-06, + "loss": 0.5335, + "step": 2253 + }, + { + "epoch": 0.5357737104825291, + "grad_norm": 1.4745858006135701, + "learning_rate": 4.665301248917981e-06, + "loss": 0.5893, + "step": 2254 + }, + { + "epoch": 0.5360114095555027, + "grad_norm": 1.218596428477352, + "learning_rate": 4.661459992976463e-06, + "loss": 0.5647, + "step": 2255 + }, + { + "epoch": 0.5362491086284763, + "grad_norm": 1.4739361155893909, + "learning_rate": 4.657618937754366e-06, + "loss": 0.5823, + "step": 2256 + }, + { + "epoch": 0.53648680770145, + "grad_norm": 1.3936001131420652, + "learning_rate": 4.653778085529043e-06, + "loss": 0.5642, + "step": 2257 + }, + { + "epoch": 0.5367245067744236, + "grad_norm": 1.2510045935432215, + "learning_rate": 4.64993743857772e-06, + "loss": 0.6043, + "step": 2258 + }, + { + "epoch": 0.5369622058473972, + "grad_norm": 1.1540662530374057, + "learning_rate": 4.646096999177511e-06, + "loss": 0.5594, + "step": 2259 + }, + { + "epoch": 0.5371999049203708, + "grad_norm": 1.9327209775414818, + "learning_rate": 4.642256769605399e-06, + "loss": 0.6051, + "step": 2260 + }, + { + "epoch": 0.5374376039933444, + "grad_norm": 2.135801994923028, + "learning_rate": 4.638416752138245e-06, + "loss": 0.5849, + "step": 2261 + }, + { + "epoch": 0.537675303066318, + "grad_norm": 1.4096311715723815, + "learning_rate": 4.634576949052786e-06, + "loss": 0.6013, + "step": 2262 + }, + { + "epoch": 0.5379130021392916, + "grad_norm": 1.9633451947950298, + "learning_rate": 4.630737362625631e-06, + "loss": 0.5789, + "step": 2263 + }, + { + "epoch": 0.5381507012122653, + "grad_norm": 1.1296826005641087, + "learning_rate": 4.626897995133257e-06, + "loss": 0.5524, + "step": 2264 + }, + { + "epoch": 0.5383884002852389, + "grad_norm": 1.7368688553076264, + "learning_rate": 4.623058848852018e-06, + "loss": 0.5658, + "step": 2265 + }, + { + "epoch": 0.5386260993582125, + "grad_norm": 1.7273788422054681, + "learning_rate": 4.61921992605813e-06, + "loss": 0.6103, + "step": 2266 + }, + { + "epoch": 0.5388637984311861, + "grad_norm": 1.270917705245557, + "learning_rate": 4.615381229027681e-06, + "loss": 0.5458, + "step": 2267 + }, + { + "epoch": 0.5391014975041597, + "grad_norm": 1.1717516643785688, + "learning_rate": 4.611542760036624e-06, + "loss": 0.5653, + "step": 2268 + }, + { + "epoch": 0.5393391965771334, + "grad_norm": 1.123181465219037, + "learning_rate": 4.6077045213607765e-06, + "loss": 0.6021, + "step": 2269 + }, + { + "epoch": 0.5395768956501069, + "grad_norm": 1.416559240834848, + "learning_rate": 4.6038665152758175e-06, + "loss": 0.5685, + "step": 2270 + }, + { + "epoch": 0.5398145947230806, + "grad_norm": 1.2581742302649097, + "learning_rate": 4.6000287440572925e-06, + "loss": 0.5946, + "step": 2271 + }, + { + "epoch": 0.5400522937960542, + "grad_norm": 1.2047057015829565, + "learning_rate": 4.596191209980604e-06, + "loss": 0.5941, + "step": 2272 + }, + { + "epoch": 0.5402899928690278, + "grad_norm": 1.1716484653495007, + "learning_rate": 4.592353915321016e-06, + "loss": 0.5747, + "step": 2273 + }, + { + "epoch": 0.5405276919420015, + "grad_norm": 1.02405075369521, + "learning_rate": 4.588516862353649e-06, + "loss": 0.5888, + "step": 2274 + }, + { + "epoch": 0.540765391014975, + "grad_norm": 1.2103318688953029, + "learning_rate": 4.584680053353481e-06, + "loss": 0.5702, + "step": 2275 + }, + { + "epoch": 0.5410030900879487, + "grad_norm": 1.3339266786603674, + "learning_rate": 4.580843490595345e-06, + "loss": 0.5687, + "step": 2276 + }, + { + "epoch": 0.5412407891609222, + "grad_norm": 1.5463835517617055, + "learning_rate": 4.577007176353931e-06, + "loss": 0.5711, + "step": 2277 + }, + { + "epoch": 0.5414784882338959, + "grad_norm": 1.1911120293501338, + "learning_rate": 4.573171112903775e-06, + "loss": 0.6069, + "step": 2278 + }, + { + "epoch": 0.5417161873068695, + "grad_norm": 1.2541768246526503, + "learning_rate": 4.569335302519271e-06, + "loss": 0.6114, + "step": 2279 + }, + { + "epoch": 0.5419538863798431, + "grad_norm": 1.160195643043021, + "learning_rate": 4.565499747474658e-06, + "loss": 0.5462, + "step": 2280 + }, + { + "epoch": 0.5421915854528168, + "grad_norm": 1.156115183446977, + "learning_rate": 4.561664450044029e-06, + "loss": 0.5904, + "step": 2281 + }, + { + "epoch": 0.5424292845257903, + "grad_norm": 1.3462301597070911, + "learning_rate": 4.5578294125013175e-06, + "loss": 0.6131, + "step": 2282 + }, + { + "epoch": 0.542666983598764, + "grad_norm": 1.0770483642931048, + "learning_rate": 4.55399463712031e-06, + "loss": 0.5735, + "step": 2283 + }, + { + "epoch": 0.5429046826717375, + "grad_norm": 1.152283771467473, + "learning_rate": 4.550160126174631e-06, + "loss": 0.5634, + "step": 2284 + }, + { + "epoch": 0.5431423817447112, + "grad_norm": 1.4009405412275475, + "learning_rate": 4.546325881937754e-06, + "loss": 0.6037, + "step": 2285 + }, + { + "epoch": 0.5433800808176849, + "grad_norm": 1.2296487682544568, + "learning_rate": 4.542491906682988e-06, + "loss": 0.5915, + "step": 2286 + }, + { + "epoch": 0.5436177798906584, + "grad_norm": 1.6241316552553882, + "learning_rate": 4.53865820268349e-06, + "loss": 0.5943, + "step": 2287 + }, + { + "epoch": 0.5438554789636321, + "grad_norm": 1.1681409501206772, + "learning_rate": 4.534824772212251e-06, + "loss": 0.5599, + "step": 2288 + }, + { + "epoch": 0.5440931780366056, + "grad_norm": 1.3340315296815275, + "learning_rate": 4.530991617542102e-06, + "loss": 0.5904, + "step": 2289 + }, + { + "epoch": 0.5443308771095793, + "grad_norm": 1.1576924401982447, + "learning_rate": 4.527158740945707e-06, + "loss": 0.5628, + "step": 2290 + }, + { + "epoch": 0.5445685761825528, + "grad_norm": 1.1931093818936278, + "learning_rate": 4.52332614469557e-06, + "loss": 0.588, + "step": 2291 + }, + { + "epoch": 0.5448062752555265, + "grad_norm": 1.2565269145374274, + "learning_rate": 4.519493831064025e-06, + "loss": 0.5664, + "step": 2292 + }, + { + "epoch": 0.5450439743285002, + "grad_norm": 1.3658838196297407, + "learning_rate": 4.515661802323244e-06, + "loss": 0.6451, + "step": 2293 + }, + { + "epoch": 0.5452816734014737, + "grad_norm": 1.267942787495178, + "learning_rate": 4.51183006074522e-06, + "loss": 0.5809, + "step": 2294 + }, + { + "epoch": 0.5455193724744474, + "grad_norm": 1.2579656915182893, + "learning_rate": 4.507998608601787e-06, + "loss": 0.5441, + "step": 2295 + }, + { + "epoch": 0.5457570715474209, + "grad_norm": 1.2091466476092048, + "learning_rate": 4.504167448164599e-06, + "loss": 0.5561, + "step": 2296 + }, + { + "epoch": 0.5459947706203946, + "grad_norm": 1.4046471962231868, + "learning_rate": 4.5003365817051434e-06, + "loss": 0.5756, + "step": 2297 + }, + { + "epoch": 0.5462324696933682, + "grad_norm": 1.0844736509917403, + "learning_rate": 4.4965060114947265e-06, + "loss": 0.6068, + "step": 2298 + }, + { + "epoch": 0.5464701687663418, + "grad_norm": 1.2442070887627619, + "learning_rate": 4.492675739804486e-06, + "loss": 0.5955, + "step": 2299 + }, + { + "epoch": 0.5467078678393155, + "grad_norm": 1.1243031330273943, + "learning_rate": 4.488845768905377e-06, + "loss": 0.5708, + "step": 2300 + }, + { + "epoch": 0.5467078678393155, + "eval_loss": 0.5884464979171753, + "eval_runtime": 291.5267, + "eval_samples_per_second": 37.314, + "eval_steps_per_second": 1.166, + "step": 2300 + }, + { + "epoch": 0.546945566912289, + "grad_norm": 1.2249379400674285, + "learning_rate": 4.4850161010681795e-06, + "loss": 0.5289, + "step": 2301 + }, + { + "epoch": 0.5471832659852627, + "grad_norm": 1.3042833132636997, + "learning_rate": 4.4811867385634916e-06, + "loss": 0.5665, + "step": 2302 + }, + { + "epoch": 0.5474209650582362, + "grad_norm": 1.1349449271505436, + "learning_rate": 4.477357683661734e-06, + "loss": 0.5319, + "step": 2303 + }, + { + "epoch": 0.5476586641312099, + "grad_norm": 1.0816774220710172, + "learning_rate": 4.4735289386331395e-06, + "loss": 0.5363, + "step": 2304 + }, + { + "epoch": 0.5478963632041836, + "grad_norm": 1.1917556290969986, + "learning_rate": 4.4697005057477634e-06, + "loss": 0.5626, + "step": 2305 + }, + { + "epoch": 0.5481340622771571, + "grad_norm": 1.464081179936916, + "learning_rate": 4.465872387275471e-06, + "loss": 0.5844, + "step": 2306 + }, + { + "epoch": 0.5483717613501308, + "grad_norm": 1.3833773351238265, + "learning_rate": 4.462044585485944e-06, + "loss": 0.5698, + "step": 2307 + }, + { + "epoch": 0.5486094604231043, + "grad_norm": 1.4797584867704208, + "learning_rate": 4.4582171026486734e-06, + "loss": 0.5991, + "step": 2308 + }, + { + "epoch": 0.548847159496078, + "grad_norm": 1.289418144036817, + "learning_rate": 4.4543899410329664e-06, + "loss": 0.6226, + "step": 2309 + }, + { + "epoch": 0.5490848585690515, + "grad_norm": 1.4022970644901238, + "learning_rate": 4.4505631029079325e-06, + "loss": 0.5048, + "step": 2310 + }, + { + "epoch": 0.5493225576420252, + "grad_norm": 1.3417748662711366, + "learning_rate": 4.446736590542497e-06, + "loss": 0.6166, + "step": 2311 + }, + { + "epoch": 0.5495602567149989, + "grad_norm": 1.2922176426897847, + "learning_rate": 4.442910406205385e-06, + "loss": 0.5801, + "step": 2312 + }, + { + "epoch": 0.5497979557879724, + "grad_norm": 1.21549480963588, + "learning_rate": 4.439084552165134e-06, + "loss": 0.5428, + "step": 2313 + }, + { + "epoch": 0.5500356548609461, + "grad_norm": 1.3038958394884521, + "learning_rate": 4.435259030690078e-06, + "loss": 0.6424, + "step": 2314 + }, + { + "epoch": 0.5502733539339196, + "grad_norm": 1.4137111992089364, + "learning_rate": 4.431433844048361e-06, + "loss": 0.6162, + "step": 2315 + }, + { + "epoch": 0.5505110530068933, + "grad_norm": 1.1428355639647698, + "learning_rate": 4.427608994507925e-06, + "loss": 0.6161, + "step": 2316 + }, + { + "epoch": 0.5507487520798668, + "grad_norm": 1.444189500178061, + "learning_rate": 4.4237844843365126e-06, + "loss": 0.5651, + "step": 2317 + }, + { + "epoch": 0.5509864511528405, + "grad_norm": 1.2272281732007935, + "learning_rate": 4.419960315801664e-06, + "loss": 0.5524, + "step": 2318 + }, + { + "epoch": 0.5512241502258142, + "grad_norm": 1.132644179053487, + "learning_rate": 4.41613649117072e-06, + "loss": 0.5387, + "step": 2319 + }, + { + "epoch": 0.5514618492987877, + "grad_norm": 1.2417936172328894, + "learning_rate": 4.4123130127108125e-06, + "loss": 0.5568, + "step": 2320 + }, + { + "epoch": 0.5516995483717614, + "grad_norm": 1.2456356981685852, + "learning_rate": 4.408489882688875e-06, + "loss": 0.601, + "step": 2321 + }, + { + "epoch": 0.5519372474447349, + "grad_norm": 1.2861789753736266, + "learning_rate": 4.404667103371626e-06, + "loss": 0.5642, + "step": 2322 + }, + { + "epoch": 0.5521749465177086, + "grad_norm": 1.2712975290942725, + "learning_rate": 4.400844677025585e-06, + "loss": 0.564, + "step": 2323 + }, + { + "epoch": 0.5524126455906821, + "grad_norm": 1.4118885628809712, + "learning_rate": 4.397022605917055e-06, + "loss": 0.5671, + "step": 2324 + }, + { + "epoch": 0.5526503446636558, + "grad_norm": 1.2265273530934986, + "learning_rate": 4.393200892312133e-06, + "loss": 0.563, + "step": 2325 + }, + { + "epoch": 0.5528880437366295, + "grad_norm": 1.0843249330003615, + "learning_rate": 4.389379538476702e-06, + "loss": 0.5773, + "step": 2326 + }, + { + "epoch": 0.553125742809603, + "grad_norm": 1.4296073006732768, + "learning_rate": 4.3855585466764305e-06, + "loss": 0.5886, + "step": 2327 + }, + { + "epoch": 0.5533634418825767, + "grad_norm": 1.4328315994204037, + "learning_rate": 4.381737919176775e-06, + "loss": 0.5851, + "step": 2328 + }, + { + "epoch": 0.5536011409555502, + "grad_norm": 1.3532438848486013, + "learning_rate": 4.377917658242975e-06, + "loss": 0.5671, + "step": 2329 + }, + { + "epoch": 0.5538388400285239, + "grad_norm": 1.3999052696829648, + "learning_rate": 4.374097766140053e-06, + "loss": 0.5801, + "step": 2330 + }, + { + "epoch": 0.5540765391014975, + "grad_norm": 1.1562919899236692, + "learning_rate": 4.370278245132811e-06, + "loss": 0.5841, + "step": 2331 + }, + { + "epoch": 0.5543142381744711, + "grad_norm": 1.296900241290606, + "learning_rate": 4.366459097485832e-06, + "loss": 0.5345, + "step": 2332 + }, + { + "epoch": 0.5545519372474448, + "grad_norm": 1.0927275800567247, + "learning_rate": 4.3626403254634795e-06, + "loss": 0.5242, + "step": 2333 + }, + { + "epoch": 0.5547896363204183, + "grad_norm": 1.0955110562549537, + "learning_rate": 4.358821931329892e-06, + "loss": 0.5162, + "step": 2334 + }, + { + "epoch": 0.555027335393392, + "grad_norm": 1.2938614448465062, + "learning_rate": 4.355003917348985e-06, + "loss": 0.5586, + "step": 2335 + }, + { + "epoch": 0.5552650344663655, + "grad_norm": 1.2330937810363811, + "learning_rate": 4.3511862857844474e-06, + "loss": 0.5748, + "step": 2336 + }, + { + "epoch": 0.5555027335393392, + "grad_norm": 1.0657631966985555, + "learning_rate": 4.347369038899744e-06, + "loss": 0.5259, + "step": 2337 + }, + { + "epoch": 0.5557404326123128, + "grad_norm": 1.155865894594616, + "learning_rate": 4.343552178958107e-06, + "loss": 0.605, + "step": 2338 + }, + { + "epoch": 0.5559781316852864, + "grad_norm": 1.1717672296819894, + "learning_rate": 4.339735708222544e-06, + "loss": 0.585, + "step": 2339 + }, + { + "epoch": 0.5562158307582601, + "grad_norm": 1.3537754263455768, + "learning_rate": 4.33591962895583e-06, + "loss": 0.5313, + "step": 2340 + }, + { + "epoch": 0.5564535298312336, + "grad_norm": 1.187267356442225, + "learning_rate": 4.332103943420507e-06, + "loss": 0.5514, + "step": 2341 + }, + { + "epoch": 0.5566912289042073, + "grad_norm": 1.1983242745973206, + "learning_rate": 4.328288653878884e-06, + "loss": 0.5804, + "step": 2342 + }, + { + "epoch": 0.5569289279771809, + "grad_norm": 1.1477383695028218, + "learning_rate": 4.324473762593037e-06, + "loss": 0.601, + "step": 2343 + }, + { + "epoch": 0.5571666270501545, + "grad_norm": 1.1804677838747155, + "learning_rate": 4.320659271824801e-06, + "loss": 0.5893, + "step": 2344 + }, + { + "epoch": 0.5574043261231281, + "grad_norm": 1.2931256404312974, + "learning_rate": 4.316845183835782e-06, + "loss": 0.5508, + "step": 2345 + }, + { + "epoch": 0.5576420251961017, + "grad_norm": 1.2705722234574635, + "learning_rate": 4.313031500887336e-06, + "loss": 0.5919, + "step": 2346 + }, + { + "epoch": 0.5578797242690754, + "grad_norm": 1.3217137474780882, + "learning_rate": 4.309218225240591e-06, + "loss": 0.5494, + "step": 2347 + }, + { + "epoch": 0.558117423342049, + "grad_norm": 1.3025469737906736, + "learning_rate": 4.305405359156422e-06, + "loss": 0.6058, + "step": 2348 + }, + { + "epoch": 0.5583551224150226, + "grad_norm": 1.2008305088303315, + "learning_rate": 4.301592904895471e-06, + "loss": 0.5488, + "step": 2349 + }, + { + "epoch": 0.5585928214879962, + "grad_norm": 1.180879787897394, + "learning_rate": 4.297780864718129e-06, + "loss": 0.5461, + "step": 2350 + }, + { + "epoch": 0.5588305205609698, + "grad_norm": 1.2714457971019049, + "learning_rate": 4.293969240884545e-06, + "loss": 0.5886, + "step": 2351 + }, + { + "epoch": 0.5590682196339434, + "grad_norm": 1.182984453968635, + "learning_rate": 4.290158035654618e-06, + "loss": 0.5662, + "step": 2352 + }, + { + "epoch": 0.559305918706917, + "grad_norm": 1.0834329833864147, + "learning_rate": 4.286347251288004e-06, + "loss": 0.5634, + "step": 2353 + }, + { + "epoch": 0.5595436177798907, + "grad_norm": 1.2493393579275152, + "learning_rate": 4.282536890044105e-06, + "loss": 0.5382, + "step": 2354 + }, + { + "epoch": 0.5597813168528643, + "grad_norm": 1.1710692287076565, + "learning_rate": 4.278726954182073e-06, + "loss": 0.5716, + "step": 2355 + }, + { + "epoch": 0.5600190159258379, + "grad_norm": 1.3591028693579115, + "learning_rate": 4.274917445960808e-06, + "loss": 0.558, + "step": 2356 + }, + { + "epoch": 0.5602567149988115, + "grad_norm": 1.2150617871286622, + "learning_rate": 4.271108367638959e-06, + "loss": 0.5976, + "step": 2357 + }, + { + "epoch": 0.5604944140717851, + "grad_norm": 1.730756166440005, + "learning_rate": 4.267299721474915e-06, + "loss": 0.5486, + "step": 2358 + }, + { + "epoch": 0.5607321131447587, + "grad_norm": 1.3198087628835637, + "learning_rate": 4.263491509726812e-06, + "loss": 0.6235, + "step": 2359 + }, + { + "epoch": 0.5609698122177323, + "grad_norm": 1.387412248340222, + "learning_rate": 4.259683734652531e-06, + "loss": 0.5912, + "step": 2360 + }, + { + "epoch": 0.561207511290706, + "grad_norm": 1.48054076104741, + "learning_rate": 4.255876398509689e-06, + "loss": 0.5948, + "step": 2361 + }, + { + "epoch": 0.5614452103636796, + "grad_norm": 1.513990398613633, + "learning_rate": 4.252069503555645e-06, + "loss": 0.6012, + "step": 2362 + }, + { + "epoch": 0.5616829094366532, + "grad_norm": 1.1725120424928228, + "learning_rate": 4.2482630520474955e-06, + "loss": 0.5815, + "step": 2363 + }, + { + "epoch": 0.5619206085096268, + "grad_norm": 1.281478999859122, + "learning_rate": 4.244457046242078e-06, + "loss": 0.591, + "step": 2364 + }, + { + "epoch": 0.5621583075826004, + "grad_norm": 1.1621646221212847, + "learning_rate": 4.240651488395958e-06, + "loss": 0.6467, + "step": 2365 + }, + { + "epoch": 0.562396006655574, + "grad_norm": 1.7838234056264586, + "learning_rate": 4.236846380765445e-06, + "loss": 0.6022, + "step": 2366 + }, + { + "epoch": 0.5626337057285477, + "grad_norm": 1.2336083025919666, + "learning_rate": 4.233041725606573e-06, + "loss": 0.6132, + "step": 2367 + }, + { + "epoch": 0.5628714048015213, + "grad_norm": 1.4359925445139397, + "learning_rate": 4.229237525175114e-06, + "loss": 0.5846, + "step": 2368 + }, + { + "epoch": 0.5631091038744949, + "grad_norm": 1.2553775340244178, + "learning_rate": 4.225433781726564e-06, + "loss": 0.645, + "step": 2369 + }, + { + "epoch": 0.5633468029474685, + "grad_norm": 1.4210027621340044, + "learning_rate": 4.221630497516157e-06, + "loss": 0.5751, + "step": 2370 + }, + { + "epoch": 0.5635845020204421, + "grad_norm": 1.1839051022159135, + "learning_rate": 4.217827674798845e-06, + "loss": 0.591, + "step": 2371 + }, + { + "epoch": 0.5638222010934157, + "grad_norm": 1.2873357157645537, + "learning_rate": 4.214025315829314e-06, + "loss": 0.5613, + "step": 2372 + }, + { + "epoch": 0.5640599001663894, + "grad_norm": 1.3880591141480418, + "learning_rate": 4.210223422861971e-06, + "loss": 0.5787, + "step": 2373 + }, + { + "epoch": 0.564297599239363, + "grad_norm": 1.1903380421373746, + "learning_rate": 4.206421998150947e-06, + "loss": 0.5393, + "step": 2374 + }, + { + "epoch": 0.5645352983123366, + "grad_norm": 1.272792482636954, + "learning_rate": 4.202621043950096e-06, + "loss": 0.5319, + "step": 2375 + }, + { + "epoch": 0.5647729973853102, + "grad_norm": 1.370082727977022, + "learning_rate": 4.198820562512994e-06, + "loss": 0.6268, + "step": 2376 + }, + { + "epoch": 0.5650106964582838, + "grad_norm": 1.3891896772143926, + "learning_rate": 4.195020556092935e-06, + "loss": 0.5751, + "step": 2377 + }, + { + "epoch": 0.5652483955312574, + "grad_norm": 1.3163256991084409, + "learning_rate": 4.191221026942934e-06, + "loss": 0.5432, + "step": 2378 + }, + { + "epoch": 0.565486094604231, + "grad_norm": 1.7404266543677402, + "learning_rate": 4.187421977315718e-06, + "loss": 0.6031, + "step": 2379 + }, + { + "epoch": 0.5657237936772047, + "grad_norm": 1.5070770387843768, + "learning_rate": 4.183623409463736e-06, + "loss": 0.5987, + "step": 2380 + }, + { + "epoch": 0.5659614927501783, + "grad_norm": 1.2255550734920237, + "learning_rate": 4.179825325639146e-06, + "loss": 0.5784, + "step": 2381 + }, + { + "epoch": 0.5661991918231519, + "grad_norm": 1.5946103845920105, + "learning_rate": 4.1760277280938226e-06, + "loss": 0.6046, + "step": 2382 + }, + { + "epoch": 0.5664368908961255, + "grad_norm": 1.7274063065061602, + "learning_rate": 4.17223061907935e-06, + "loss": 0.5861, + "step": 2383 + }, + { + "epoch": 0.5666745899690991, + "grad_norm": 1.1533004889404976, + "learning_rate": 4.168434000847024e-06, + "loss": 0.6042, + "step": 2384 + }, + { + "epoch": 0.5669122890420727, + "grad_norm": 1.7447643019729875, + "learning_rate": 4.164637875647848e-06, + "loss": 0.5813, + "step": 2385 + }, + { + "epoch": 0.5671499881150464, + "grad_norm": 1.3682135270259672, + "learning_rate": 4.160842245732536e-06, + "loss": 0.6235, + "step": 2386 + }, + { + "epoch": 0.56738768718802, + "grad_norm": 1.1828539468663462, + "learning_rate": 4.157047113351504e-06, + "loss": 0.5717, + "step": 2387 + }, + { + "epoch": 0.5676253862609936, + "grad_norm": 1.2664445187947393, + "learning_rate": 4.1532524807548776e-06, + "loss": 0.5909, + "step": 2388 + }, + { + "epoch": 0.5678630853339672, + "grad_norm": 1.0874647430032285, + "learning_rate": 4.14945835019248e-06, + "loss": 0.5779, + "step": 2389 + }, + { + "epoch": 0.5681007844069408, + "grad_norm": 1.2130681205342073, + "learning_rate": 4.145664723913846e-06, + "loss": 0.5468, + "step": 2390 + }, + { + "epoch": 0.5683384834799144, + "grad_norm": 1.3483972138495668, + "learning_rate": 4.141871604168201e-06, + "loss": 0.5974, + "step": 2391 + }, + { + "epoch": 0.568576182552888, + "grad_norm": 1.6916451666398267, + "learning_rate": 4.1380789932044794e-06, + "loss": 0.5918, + "step": 2392 + }, + { + "epoch": 0.5688138816258617, + "grad_norm": 1.4666132249061825, + "learning_rate": 4.134286893271306e-06, + "loss": 0.6025, + "step": 2393 + }, + { + "epoch": 0.5690515806988353, + "grad_norm": 1.2230330635267412, + "learning_rate": 4.130495306617009e-06, + "loss": 0.5972, + "step": 2394 + }, + { + "epoch": 0.5692892797718089, + "grad_norm": 1.1079551796556273, + "learning_rate": 4.126704235489606e-06, + "loss": 0.5372, + "step": 2395 + }, + { + "epoch": 0.5695269788447825, + "grad_norm": 1.0849376335577166, + "learning_rate": 4.122913682136816e-06, + "loss": 0.5644, + "step": 2396 + }, + { + "epoch": 0.5697646779177561, + "grad_norm": 1.2567045140864777, + "learning_rate": 4.119123648806046e-06, + "loss": 0.6039, + "step": 2397 + }, + { + "epoch": 0.5700023769907298, + "grad_norm": 1.2235855171374395, + "learning_rate": 4.115334137744397e-06, + "loss": 0.5487, + "step": 2398 + }, + { + "epoch": 0.5702400760637033, + "grad_norm": 1.2201731848946595, + "learning_rate": 4.111545151198657e-06, + "loss": 0.6077, + "step": 2399 + }, + { + "epoch": 0.570477775136677, + "grad_norm": 1.2291801754113052, + "learning_rate": 4.107756691415308e-06, + "loss": 0.5443, + "step": 2400 + }, + { + "epoch": 0.570477775136677, + "eval_loss": 0.5845754742622375, + "eval_runtime": 292.1956, + "eval_samples_per_second": 37.228, + "eval_steps_per_second": 1.164, + "step": 2400 + }, + { + "epoch": 0.5707154742096506, + "grad_norm": 1.2575370754709303, + "learning_rate": 4.103968760640516e-06, + "loss": 0.5085, + "step": 2401 + }, + { + "epoch": 0.5709531732826242, + "grad_norm": 2.2272823493553657, + "learning_rate": 4.100181361120136e-06, + "loss": 0.6052, + "step": 2402 + }, + { + "epoch": 0.5711908723555978, + "grad_norm": 1.3009165954322706, + "learning_rate": 4.0963944950997046e-06, + "loss": 0.571, + "step": 2403 + }, + { + "epoch": 0.5714285714285714, + "grad_norm": 1.5723987593453015, + "learning_rate": 4.092608164824446e-06, + "loss": 0.6669, + "step": 2404 + }, + { + "epoch": 0.5716662705015451, + "grad_norm": 1.4687745231624487, + "learning_rate": 4.088822372539263e-06, + "loss": 0.5856, + "step": 2405 + }, + { + "epoch": 0.5719039695745186, + "grad_norm": 1.7792845077549508, + "learning_rate": 4.085037120488742e-06, + "loss": 0.5625, + "step": 2406 + }, + { + "epoch": 0.5721416686474923, + "grad_norm": 1.17907923343007, + "learning_rate": 4.081252410917148e-06, + "loss": 0.5586, + "step": 2407 + }, + { + "epoch": 0.5723793677204659, + "grad_norm": 1.3182107284026219, + "learning_rate": 4.0774682460684264e-06, + "loss": 0.5649, + "step": 2408 + }, + { + "epoch": 0.5726170667934395, + "grad_norm": 1.4352563618897276, + "learning_rate": 4.073684628186195e-06, + "loss": 0.5818, + "step": 2409 + }, + { + "epoch": 0.5728547658664132, + "grad_norm": 1.4817748949865412, + "learning_rate": 4.0699015595137535e-06, + "loss": 0.585, + "step": 2410 + }, + { + "epoch": 0.5730924649393867, + "grad_norm": 1.6508377599283748, + "learning_rate": 4.06611904229407e-06, + "loss": 0.6075, + "step": 2411 + }, + { + "epoch": 0.5733301640123604, + "grad_norm": 1.287756713356248, + "learning_rate": 4.062337078769789e-06, + "loss": 0.5963, + "step": 2412 + }, + { + "epoch": 0.5735678630853339, + "grad_norm": 2.312950591860136, + "learning_rate": 4.058555671183227e-06, + "loss": 0.6219, + "step": 2413 + }, + { + "epoch": 0.5738055621583076, + "grad_norm": 1.190975091619916, + "learning_rate": 4.05477482177637e-06, + "loss": 0.5565, + "step": 2414 + }, + { + "epoch": 0.5740432612312812, + "grad_norm": 1.6493713203746172, + "learning_rate": 4.050994532790871e-06, + "loss": 0.5788, + "step": 2415 + }, + { + "epoch": 0.5742809603042548, + "grad_norm": 1.406866364043468, + "learning_rate": 4.047214806468056e-06, + "loss": 0.5935, + "step": 2416 + }, + { + "epoch": 0.5745186593772285, + "grad_norm": 1.1435798162930468, + "learning_rate": 4.043435645048911e-06, + "loss": 0.5287, + "step": 2417 + }, + { + "epoch": 0.574756358450202, + "grad_norm": 1.4203792195330989, + "learning_rate": 4.039657050774093e-06, + "loss": 0.6088, + "step": 2418 + }, + { + "epoch": 0.5749940575231757, + "grad_norm": 1.365812453004828, + "learning_rate": 4.035879025883916e-06, + "loss": 0.6133, + "step": 2419 + }, + { + "epoch": 0.5752317565961492, + "grad_norm": 1.1760445033070415, + "learning_rate": 4.0321015726183655e-06, + "loss": 0.6085, + "step": 2420 + }, + { + "epoch": 0.5754694556691229, + "grad_norm": 1.2084920649094504, + "learning_rate": 4.028324693217078e-06, + "loss": 0.5524, + "step": 2421 + }, + { + "epoch": 0.5757071547420965, + "grad_norm": 1.124236500960957, + "learning_rate": 4.02454838991936e-06, + "loss": 0.5888, + "step": 2422 + }, + { + "epoch": 0.5759448538150701, + "grad_norm": 1.1582583802001294, + "learning_rate": 4.020772664964166e-06, + "loss": 0.5143, + "step": 2423 + }, + { + "epoch": 0.5761825528880438, + "grad_norm": 1.484946430426738, + "learning_rate": 4.016997520590115e-06, + "loss": 0.5636, + "step": 2424 + }, + { + "epoch": 0.5764202519610173, + "grad_norm": 1.7017967613754774, + "learning_rate": 4.013222959035481e-06, + "loss": 0.5415, + "step": 2425 + }, + { + "epoch": 0.576657951033991, + "grad_norm": 1.9579523203632996, + "learning_rate": 4.009448982538188e-06, + "loss": 0.6076, + "step": 2426 + }, + { + "epoch": 0.5768956501069645, + "grad_norm": 1.2060268357586974, + "learning_rate": 4.005675593335818e-06, + "loss": 0.5343, + "step": 2427 + }, + { + "epoch": 0.5771333491799382, + "grad_norm": 1.162050959402707, + "learning_rate": 4.001902793665602e-06, + "loss": 0.5368, + "step": 2428 + }, + { + "epoch": 0.5773710482529119, + "grad_norm": 1.1701120970325076, + "learning_rate": 3.998130585764424e-06, + "loss": 0.6006, + "step": 2429 + }, + { + "epoch": 0.5776087473258854, + "grad_norm": 1.6520504961774214, + "learning_rate": 3.9943589718688125e-06, + "loss": 0.521, + "step": 2430 + }, + { + "epoch": 0.5778464463988591, + "grad_norm": 1.1866369043379053, + "learning_rate": 3.99058795421495e-06, + "loss": 0.5774, + "step": 2431 + }, + { + "epoch": 0.5780841454718326, + "grad_norm": 1.354965308978671, + "learning_rate": 3.98681753503866e-06, + "loss": 0.5914, + "step": 2432 + }, + { + "epoch": 0.5783218445448063, + "grad_norm": 1.1508822162109356, + "learning_rate": 3.983047716575415e-06, + "loss": 0.5686, + "step": 2433 + }, + { + "epoch": 0.5785595436177798, + "grad_norm": 1.4665228786782087, + "learning_rate": 3.979278501060328e-06, + "loss": 0.5894, + "step": 2434 + }, + { + "epoch": 0.5787972426907535, + "grad_norm": 1.2227534894105774, + "learning_rate": 3.975509890728159e-06, + "loss": 0.6163, + "step": 2435 + }, + { + "epoch": 0.5790349417637272, + "grad_norm": 1.4804228672591733, + "learning_rate": 3.971741887813304e-06, + "loss": 0.5809, + "step": 2436 + }, + { + "epoch": 0.5792726408367007, + "grad_norm": 1.343684336853829, + "learning_rate": 3.967974494549803e-06, + "loss": 0.563, + "step": 2437 + }, + { + "epoch": 0.5795103399096744, + "grad_norm": 1.221990622124212, + "learning_rate": 3.964207713171331e-06, + "loss": 0.6032, + "step": 2438 + }, + { + "epoch": 0.5797480389826479, + "grad_norm": 1.2429424280821177, + "learning_rate": 3.960441545911205e-06, + "loss": 0.5413, + "step": 2439 + }, + { + "epoch": 0.5799857380556216, + "grad_norm": 1.3575768393399668, + "learning_rate": 3.956675995002372e-06, + "loss": 0.5568, + "step": 2440 + }, + { + "epoch": 0.5802234371285951, + "grad_norm": 1.194258358652641, + "learning_rate": 3.952911062677418e-06, + "loss": 0.5671, + "step": 2441 + }, + { + "epoch": 0.5804611362015688, + "grad_norm": 1.4384215169902186, + "learning_rate": 3.949146751168562e-06, + "loss": 0.5641, + "step": 2442 + }, + { + "epoch": 0.5806988352745425, + "grad_norm": 1.465024213260641, + "learning_rate": 3.945383062707652e-06, + "loss": 0.5908, + "step": 2443 + }, + { + "epoch": 0.580936534347516, + "grad_norm": 1.3956030080953479, + "learning_rate": 3.9416199995261674e-06, + "loss": 0.5909, + "step": 2444 + }, + { + "epoch": 0.5811742334204897, + "grad_norm": 1.3257046654034408, + "learning_rate": 3.9378575638552216e-06, + "loss": 0.6316, + "step": 2445 + }, + { + "epoch": 0.5814119324934632, + "grad_norm": 1.038511152700517, + "learning_rate": 3.934095757925549e-06, + "loss": 0.5626, + "step": 2446 + }, + { + "epoch": 0.5816496315664369, + "grad_norm": 1.0916669793825955, + "learning_rate": 3.930334583967514e-06, + "loss": 0.5538, + "step": 2447 + }, + { + "epoch": 0.5818873306394106, + "grad_norm": 1.3255143715093505, + "learning_rate": 3.926574044211107e-06, + "loss": 0.6079, + "step": 2448 + }, + { + "epoch": 0.5821250297123841, + "grad_norm": 1.352561142022113, + "learning_rate": 3.922814140885942e-06, + "loss": 0.6015, + "step": 2449 + }, + { + "epoch": 0.5823627287853578, + "grad_norm": 1.0921700951540478, + "learning_rate": 3.919054876221253e-06, + "loss": 0.5475, + "step": 2450 + }, + { + "epoch": 0.5826004278583313, + "grad_norm": 1.1235289110403237, + "learning_rate": 3.9152962524459e-06, + "loss": 0.5509, + "step": 2451 + }, + { + "epoch": 0.582838126931305, + "grad_norm": 1.270719599616304, + "learning_rate": 3.911538271788359e-06, + "loss": 0.5759, + "step": 2452 + }, + { + "epoch": 0.5830758260042785, + "grad_norm": 1.2988897982994434, + "learning_rate": 3.9077809364767275e-06, + "loss": 0.5721, + "step": 2453 + }, + { + "epoch": 0.5833135250772522, + "grad_norm": 1.2116835466914948, + "learning_rate": 3.904024248738716e-06, + "loss": 0.5803, + "step": 2454 + }, + { + "epoch": 0.5835512241502259, + "grad_norm": 1.2241952958672182, + "learning_rate": 3.9002682108016585e-06, + "loss": 0.5068, + "step": 2455 + }, + { + "epoch": 0.5837889232231994, + "grad_norm": 1.0661873471802996, + "learning_rate": 3.8965128248924956e-06, + "loss": 0.5548, + "step": 2456 + }, + { + "epoch": 0.5840266222961731, + "grad_norm": 1.5471429433477648, + "learning_rate": 3.892758093237788e-06, + "loss": 0.6068, + "step": 2457 + }, + { + "epoch": 0.5842643213691466, + "grad_norm": 1.0995242624332977, + "learning_rate": 3.889004018063702e-06, + "loss": 0.5568, + "step": 2458 + }, + { + "epoch": 0.5845020204421203, + "grad_norm": 1.21876632907325, + "learning_rate": 3.885250601596023e-06, + "loss": 0.5598, + "step": 2459 + }, + { + "epoch": 0.5847397195150938, + "grad_norm": 1.290934492650233, + "learning_rate": 3.881497846060137e-06, + "loss": 0.5847, + "step": 2460 + }, + { + "epoch": 0.5849774185880675, + "grad_norm": 1.1732107083099617, + "learning_rate": 3.8777457536810446e-06, + "loss": 0.5543, + "step": 2461 + }, + { + "epoch": 0.5852151176610412, + "grad_norm": 1.5416267804123645, + "learning_rate": 3.873994326683349e-06, + "loss": 0.5734, + "step": 2462 + }, + { + "epoch": 0.5854528167340147, + "grad_norm": 1.2628004375802615, + "learning_rate": 3.870243567291263e-06, + "loss": 0.5781, + "step": 2463 + }, + { + "epoch": 0.5856905158069884, + "grad_norm": 1.0752132697973649, + "learning_rate": 3.866493477728599e-06, + "loss": 0.5566, + "step": 2464 + }, + { + "epoch": 0.5859282148799619, + "grad_norm": 1.2339518385240607, + "learning_rate": 3.862744060218777e-06, + "loss": 0.5636, + "step": 2465 + }, + { + "epoch": 0.5861659139529356, + "grad_norm": 1.1907297212423957, + "learning_rate": 3.858995316984814e-06, + "loss": 0.5826, + "step": 2466 + }, + { + "epoch": 0.5864036130259092, + "grad_norm": 1.3199352922243612, + "learning_rate": 3.855247250249331e-06, + "loss": 0.6111, + "step": 2467 + }, + { + "epoch": 0.5866413120988828, + "grad_norm": 1.254698028956663, + "learning_rate": 3.851499862234544e-06, + "loss": 0.5937, + "step": 2468 + }, + { + "epoch": 0.5868790111718565, + "grad_norm": 1.5393517364916771, + "learning_rate": 3.847753155162272e-06, + "loss": 0.5567, + "step": 2469 + }, + { + "epoch": 0.58711671024483, + "grad_norm": 1.2091878289601288, + "learning_rate": 3.844007131253925e-06, + "loss": 0.5887, + "step": 2470 + }, + { + "epoch": 0.5873544093178037, + "grad_norm": 2.3052387687838234, + "learning_rate": 3.840261792730511e-06, + "loss": 0.6074, + "step": 2471 + }, + { + "epoch": 0.5875921083907772, + "grad_norm": 1.0342578417696986, + "learning_rate": 3.836517141812629e-06, + "loss": 0.564, + "step": 2472 + }, + { + "epoch": 0.5878298074637509, + "grad_norm": 1.1490959100387326, + "learning_rate": 3.832773180720475e-06, + "loss": 0.5662, + "step": 2473 + }, + { + "epoch": 0.5880675065367245, + "grad_norm": 1.3928526779013357, + "learning_rate": 3.82902991167383e-06, + "loss": 0.5483, + "step": 2474 + }, + { + "epoch": 0.5883052056096981, + "grad_norm": 1.4410161085658486, + "learning_rate": 3.82528733689207e-06, + "loss": 0.5864, + "step": 2475 + }, + { + "epoch": 0.5885429046826718, + "grad_norm": 1.4358058123057125, + "learning_rate": 3.821545458594155e-06, + "loss": 0.5605, + "step": 2476 + }, + { + "epoch": 0.5887806037556453, + "grad_norm": 1.0643575342018774, + "learning_rate": 3.8178042789986355e-06, + "loss": 0.5868, + "step": 2477 + }, + { + "epoch": 0.589018302828619, + "grad_norm": 1.444403611476714, + "learning_rate": 3.8140638003236462e-06, + "loss": 0.5684, + "step": 2478 + }, + { + "epoch": 0.5892560019015926, + "grad_norm": 1.2333898834367056, + "learning_rate": 3.8103240247869077e-06, + "loss": 0.5805, + "step": 2479 + }, + { + "epoch": 0.5894937009745662, + "grad_norm": 1.3459903379524165, + "learning_rate": 3.80658495460572e-06, + "loss": 0.596, + "step": 2480 + }, + { + "epoch": 0.5897314000475398, + "grad_norm": 1.5128619221865283, + "learning_rate": 3.802846591996969e-06, + "loss": 0.572, + "step": 2481 + }, + { + "epoch": 0.5899690991205134, + "grad_norm": 1.374074500002436, + "learning_rate": 3.7991089391771185e-06, + "loss": 0.5801, + "step": 2482 + }, + { + "epoch": 0.5902067981934871, + "grad_norm": 1.2385266230984033, + "learning_rate": 3.795371998362214e-06, + "loss": 0.5892, + "step": 2483 + }, + { + "epoch": 0.5904444972664606, + "grad_norm": 1.4180201993383978, + "learning_rate": 3.791635771767875e-06, + "loss": 0.5518, + "step": 2484 + }, + { + "epoch": 0.5906821963394343, + "grad_norm": 1.3196408081228137, + "learning_rate": 3.7879002616093015e-06, + "loss": 0.6047, + "step": 2485 + }, + { + "epoch": 0.5909198954124079, + "grad_norm": 1.2479432419482543, + "learning_rate": 3.784165470101264e-06, + "loss": 0.5988, + "step": 2486 + }, + { + "epoch": 0.5911575944853815, + "grad_norm": 1.1929667931978771, + "learning_rate": 3.7804313994581143e-06, + "loss": 0.5563, + "step": 2487 + }, + { + "epoch": 0.5913952935583551, + "grad_norm": 1.1951207537761355, + "learning_rate": 3.7766980518937705e-06, + "loss": 0.566, + "step": 2488 + }, + { + "epoch": 0.5916329926313287, + "grad_norm": 1.4678727627119628, + "learning_rate": 3.7729654296217234e-06, + "loss": 0.628, + "step": 2489 + }, + { + "epoch": 0.5918706917043024, + "grad_norm": 1.2013121554266974, + "learning_rate": 3.769233534855035e-06, + "loss": 0.5535, + "step": 2490 + }, + { + "epoch": 0.592108390777276, + "grad_norm": 1.4690866004859746, + "learning_rate": 3.765502369806334e-06, + "loss": 0.5661, + "step": 2491 + }, + { + "epoch": 0.5923460898502496, + "grad_norm": 1.260433609380036, + "learning_rate": 3.7617719366878204e-06, + "loss": 0.5807, + "step": 2492 + }, + { + "epoch": 0.5925837889232232, + "grad_norm": 1.3702109794487576, + "learning_rate": 3.758042237711254e-06, + "loss": 0.5882, + "step": 2493 + }, + { + "epoch": 0.5928214879961968, + "grad_norm": 1.2363284523076075, + "learning_rate": 3.7543132750879663e-06, + "loss": 0.5792, + "step": 2494 + }, + { + "epoch": 0.5930591870691704, + "grad_norm": 1.2183625814803591, + "learning_rate": 3.7505850510288455e-06, + "loss": 0.6148, + "step": 2495 + }, + { + "epoch": 0.593296886142144, + "grad_norm": 1.2590547280541002, + "learning_rate": 3.7468575677443484e-06, + "loss": 0.6022, + "step": 2496 + }, + { + "epoch": 0.5935345852151177, + "grad_norm": 1.5238931718893507, + "learning_rate": 3.743130827444487e-06, + "loss": 0.5634, + "step": 2497 + }, + { + "epoch": 0.5937722842880913, + "grad_norm": 1.4000954815218085, + "learning_rate": 3.739404832338836e-06, + "loss": 0.5701, + "step": 2498 + }, + { + "epoch": 0.5940099833610649, + "grad_norm": 1.113860658214439, + "learning_rate": 3.7356795846365268e-06, + "loss": 0.5794, + "step": 2499 + }, + { + "epoch": 0.5942476824340385, + "grad_norm": 1.3703926984019068, + "learning_rate": 3.7319550865462506e-06, + "loss": 0.5901, + "step": 2500 + }, + { + "epoch": 0.5942476824340385, + "eval_loss": 0.5812686681747437, + "eval_runtime": 297.2562, + "eval_samples_per_second": 36.595, + "eval_steps_per_second": 1.144, + "step": 2500 + }, + { + "epoch": 0.5944853815070121, + "grad_norm": 1.5145565519778905, + "learning_rate": 3.7282313402762487e-06, + "loss": 0.611, + "step": 2501 + }, + { + "epoch": 0.5947230805799857, + "grad_norm": 1.273982976557611, + "learning_rate": 3.7245083480343225e-06, + "loss": 0.5308, + "step": 2502 + }, + { + "epoch": 0.5949607796529593, + "grad_norm": 1.2811785267606606, + "learning_rate": 3.720786112027822e-06, + "loss": 0.5468, + "step": 2503 + }, + { + "epoch": 0.595198478725933, + "grad_norm": 1.239145335408901, + "learning_rate": 3.7170646344636516e-06, + "loss": 0.5724, + "step": 2504 + }, + { + "epoch": 0.5954361777989066, + "grad_norm": 1.2343027476829054, + "learning_rate": 3.7133439175482633e-06, + "loss": 0.5424, + "step": 2505 + }, + { + "epoch": 0.5956738768718802, + "grad_norm": 1.2167322202879423, + "learning_rate": 3.7096239634876625e-06, + "loss": 0.5961, + "step": 2506 + }, + { + "epoch": 0.5959115759448538, + "grad_norm": 1.3785514036504354, + "learning_rate": 3.705904774487396e-06, + "loss": 0.6081, + "step": 2507 + }, + { + "epoch": 0.5961492750178274, + "grad_norm": 1.4169020442944331, + "learning_rate": 3.7021863527525637e-06, + "loss": 0.5792, + "step": 2508 + }, + { + "epoch": 0.596386974090801, + "grad_norm": 1.1685458117090786, + "learning_rate": 3.6984687004878052e-06, + "loss": 0.5782, + "step": 2509 + }, + { + "epoch": 0.5966246731637747, + "grad_norm": 1.5071540219238568, + "learning_rate": 3.694751819897308e-06, + "loss": 0.5755, + "step": 2510 + }, + { + "epoch": 0.5968623722367483, + "grad_norm": 1.2481201730834228, + "learning_rate": 3.6910357131847986e-06, + "loss": 0.5798, + "step": 2511 + }, + { + "epoch": 0.5971000713097219, + "grad_norm": 1.2887407560275643, + "learning_rate": 3.6873203825535473e-06, + "loss": 0.548, + "step": 2512 + }, + { + "epoch": 0.5973377703826955, + "grad_norm": 1.2499989715728836, + "learning_rate": 3.683605830206362e-06, + "loss": 0.5708, + "step": 2513 + }, + { + "epoch": 0.5975754694556691, + "grad_norm": 1.3028097504973042, + "learning_rate": 3.6798920583455923e-06, + "loss": 0.5788, + "step": 2514 + }, + { + "epoch": 0.5978131685286427, + "grad_norm": 1.4042837757669204, + "learning_rate": 3.6761790691731207e-06, + "loss": 0.5873, + "step": 2515 + }, + { + "epoch": 0.5980508676016163, + "grad_norm": 1.2686112022573395, + "learning_rate": 3.6724668648903706e-06, + "loss": 0.6455, + "step": 2516 + }, + { + "epoch": 0.59828856667459, + "grad_norm": 1.2059230245227894, + "learning_rate": 3.6687554476982954e-06, + "loss": 0.5738, + "step": 2517 + }, + { + "epoch": 0.5985262657475636, + "grad_norm": 1.4009152417685031, + "learning_rate": 3.6650448197973855e-06, + "loss": 0.5537, + "step": 2518 + }, + { + "epoch": 0.5987639648205372, + "grad_norm": 1.1008205282867807, + "learning_rate": 3.6613349833876607e-06, + "loss": 0.5429, + "step": 2519 + }, + { + "epoch": 0.5990016638935108, + "grad_norm": 1.3936960601421287, + "learning_rate": 3.6576259406686732e-06, + "loss": 0.6263, + "step": 2520 + }, + { + "epoch": 0.5992393629664844, + "grad_norm": 1.1727616406033767, + "learning_rate": 3.6539176938395037e-06, + "loss": 0.5961, + "step": 2521 + }, + { + "epoch": 0.599477062039458, + "grad_norm": 1.558198090947517, + "learning_rate": 3.650210245098762e-06, + "loss": 0.5894, + "step": 2522 + }, + { + "epoch": 0.5997147611124317, + "grad_norm": 1.1086289906413878, + "learning_rate": 3.646503596644583e-06, + "loss": 0.5571, + "step": 2523 + }, + { + "epoch": 0.5999524601854053, + "grad_norm": 1.2755033748575877, + "learning_rate": 3.6427977506746293e-06, + "loss": 0.5457, + "step": 2524 + }, + { + "epoch": 0.6001901592583789, + "grad_norm": 1.7677420481997623, + "learning_rate": 3.639092709386085e-06, + "loss": 0.6142, + "step": 2525 + }, + { + "epoch": 0.6004278583313525, + "grad_norm": 1.5098303077764599, + "learning_rate": 3.635388474975662e-06, + "loss": 0.5131, + "step": 2526 + }, + { + "epoch": 0.6006655574043261, + "grad_norm": 1.501046205378393, + "learning_rate": 3.6316850496395863e-06, + "loss": 0.5421, + "step": 2527 + }, + { + "epoch": 0.6009032564772997, + "grad_norm": 1.4030002398476162, + "learning_rate": 3.627982435573611e-06, + "loss": 0.5667, + "step": 2528 + }, + { + "epoch": 0.6011409555502734, + "grad_norm": 1.3084990134825527, + "learning_rate": 3.624280634973004e-06, + "loss": 0.5854, + "step": 2529 + }, + { + "epoch": 0.601378654623247, + "grad_norm": 1.2193211606218828, + "learning_rate": 3.620579650032555e-06, + "loss": 0.5686, + "step": 2530 + }, + { + "epoch": 0.6016163536962206, + "grad_norm": 1.287683348984391, + "learning_rate": 3.616879482946565e-06, + "loss": 0.5729, + "step": 2531 + }, + { + "epoch": 0.6018540527691942, + "grad_norm": 1.198826596589926, + "learning_rate": 3.613180135908855e-06, + "loss": 0.5539, + "step": 2532 + }, + { + "epoch": 0.6020917518421678, + "grad_norm": 1.3443696277197834, + "learning_rate": 3.609481611112755e-06, + "loss": 0.6139, + "step": 2533 + }, + { + "epoch": 0.6023294509151415, + "grad_norm": 1.1726506141617838, + "learning_rate": 3.605783910751113e-06, + "loss": 0.5687, + "step": 2534 + }, + { + "epoch": 0.602567149988115, + "grad_norm": 1.201292519499824, + "learning_rate": 3.602087037016281e-06, + "loss": 0.5747, + "step": 2535 + }, + { + "epoch": 0.6028048490610887, + "grad_norm": 1.3048500715000564, + "learning_rate": 3.598390992100129e-06, + "loss": 0.4945, + "step": 2536 + }, + { + "epoch": 0.6030425481340623, + "grad_norm": 1.2687627213211985, + "learning_rate": 3.5946957781940296e-06, + "loss": 0.5922, + "step": 2537 + }, + { + "epoch": 0.6032802472070359, + "grad_norm": 1.387682463666809, + "learning_rate": 3.591001397488867e-06, + "loss": 0.61, + "step": 2538 + }, + { + "epoch": 0.6035179462800095, + "grad_norm": 1.2336752051596673, + "learning_rate": 3.587307852175025e-06, + "loss": 0.5756, + "step": 2539 + }, + { + "epoch": 0.6037556453529831, + "grad_norm": 1.274339710228977, + "learning_rate": 3.5836151444424004e-06, + "loss": 0.5723, + "step": 2540 + }, + { + "epoch": 0.6039933444259568, + "grad_norm": 1.4281835557934643, + "learning_rate": 3.579923276480387e-06, + "loss": 0.5647, + "step": 2541 + }, + { + "epoch": 0.6042310434989303, + "grad_norm": 1.2012654505314928, + "learning_rate": 3.5762322504778846e-06, + "loss": 0.5549, + "step": 2542 + }, + { + "epoch": 0.604468742571904, + "grad_norm": 1.6580476076640736, + "learning_rate": 3.5725420686232903e-06, + "loss": 0.5466, + "step": 2543 + }, + { + "epoch": 0.6047064416448776, + "grad_norm": 1.6073491062244945, + "learning_rate": 3.568852733104504e-06, + "loss": 0.5434, + "step": 2544 + }, + { + "epoch": 0.6049441407178512, + "grad_norm": 1.214216838484165, + "learning_rate": 3.5651642461089207e-06, + "loss": 0.5962, + "step": 2545 + }, + { + "epoch": 0.6051818397908248, + "grad_norm": 1.1923096756821894, + "learning_rate": 3.561476609823437e-06, + "loss": 0.5621, + "step": 2546 + }, + { + "epoch": 0.6054195388637984, + "grad_norm": 1.138197441968531, + "learning_rate": 3.557789826434439e-06, + "loss": 0.5241, + "step": 2547 + }, + { + "epoch": 0.6056572379367721, + "grad_norm": 1.248456573777256, + "learning_rate": 3.5541038981278126e-06, + "loss": 0.5337, + "step": 2548 + }, + { + "epoch": 0.6058949370097456, + "grad_norm": 1.4036360489061135, + "learning_rate": 3.550418827088932e-06, + "loss": 0.576, + "step": 2549 + }, + { + "epoch": 0.6061326360827193, + "grad_norm": 1.2077565402158557, + "learning_rate": 3.5467346155026685e-06, + "loss": 0.5619, + "step": 2550 + }, + { + "epoch": 0.6063703351556929, + "grad_norm": 2.0508340823558178, + "learning_rate": 3.5430512655533774e-06, + "loss": 0.6031, + "step": 2551 + }, + { + "epoch": 0.6066080342286665, + "grad_norm": 1.3137964396175972, + "learning_rate": 3.5393687794249093e-06, + "loss": 0.5719, + "step": 2552 + }, + { + "epoch": 0.6068457333016402, + "grad_norm": 1.3219561673197218, + "learning_rate": 3.535687159300598e-06, + "loss": 0.5923, + "step": 2553 + }, + { + "epoch": 0.6070834323746137, + "grad_norm": 1.2630415827152688, + "learning_rate": 3.5320064073632677e-06, + "loss": 0.5451, + "step": 2554 + }, + { + "epoch": 0.6073211314475874, + "grad_norm": 1.2480153638326872, + "learning_rate": 3.528326525795224e-06, + "loss": 0.5446, + "step": 2555 + }, + { + "epoch": 0.6075588305205609, + "grad_norm": 1.3357359779592615, + "learning_rate": 3.5246475167782603e-06, + "loss": 0.5527, + "step": 2556 + }, + { + "epoch": 0.6077965295935346, + "grad_norm": 1.1961217898828143, + "learning_rate": 3.5209693824936486e-06, + "loss": 0.5216, + "step": 2557 + }, + { + "epoch": 0.6080342286665082, + "grad_norm": 1.2023041911223598, + "learning_rate": 3.517292125122146e-06, + "loss": 0.5676, + "step": 2558 + }, + { + "epoch": 0.6082719277394818, + "grad_norm": 1.2490004118515352, + "learning_rate": 3.513615746843987e-06, + "loss": 0.5125, + "step": 2559 + }, + { + "epoch": 0.6085096268124555, + "grad_norm": 1.2562459450504126, + "learning_rate": 3.5099402498388877e-06, + "loss": 0.5466, + "step": 2560 + }, + { + "epoch": 0.608747325885429, + "grad_norm": 1.147741829891095, + "learning_rate": 3.5062656362860376e-06, + "loss": 0.5217, + "step": 2561 + }, + { + "epoch": 0.6089850249584027, + "grad_norm": 1.170667126015528, + "learning_rate": 3.502591908364107e-06, + "loss": 0.526, + "step": 2562 + }, + { + "epoch": 0.6092227240313762, + "grad_norm": 1.2361335810713434, + "learning_rate": 3.498919068251237e-06, + "loss": 0.5325, + "step": 2563 + }, + { + "epoch": 0.6094604231043499, + "grad_norm": 1.308121184277646, + "learning_rate": 3.495247118125046e-06, + "loss": 0.5567, + "step": 2564 + }, + { + "epoch": 0.6096981221773236, + "grad_norm": 1.2411576107449036, + "learning_rate": 3.4915760601626216e-06, + "loss": 0.591, + "step": 2565 + }, + { + "epoch": 0.6099358212502971, + "grad_norm": 1.355583339509102, + "learning_rate": 3.4879058965405242e-06, + "loss": 0.5551, + "step": 2566 + }, + { + "epoch": 0.6101735203232708, + "grad_norm": 1.1922505199590347, + "learning_rate": 3.484236629434783e-06, + "loss": 0.5504, + "step": 2567 + }, + { + "epoch": 0.6104112193962443, + "grad_norm": 1.352694574492745, + "learning_rate": 3.480568261020898e-06, + "loss": 0.5652, + "step": 2568 + }, + { + "epoch": 0.610648918469218, + "grad_norm": 1.2719215058407691, + "learning_rate": 3.476900793473832e-06, + "loss": 0.5469, + "step": 2569 + }, + { + "epoch": 0.6108866175421915, + "grad_norm": 1.256499602335687, + "learning_rate": 3.4732342289680186e-06, + "loss": 0.5241, + "step": 2570 + }, + { + "epoch": 0.6111243166151652, + "grad_norm": 1.3042572703658355, + "learning_rate": 3.469568569677352e-06, + "loss": 0.5595, + "step": 2571 + }, + { + "epoch": 0.6113620156881389, + "grad_norm": 1.240018703634632, + "learning_rate": 3.4659038177751918e-06, + "loss": 0.5513, + "step": 2572 + }, + { + "epoch": 0.6115997147611124, + "grad_norm": 1.256262966533243, + "learning_rate": 3.462239975434359e-06, + "loss": 0.572, + "step": 2573 + }, + { + "epoch": 0.6118374138340861, + "grad_norm": 1.3404663946477728, + "learning_rate": 3.458577044827136e-06, + "loss": 0.5862, + "step": 2574 + }, + { + "epoch": 0.6120751129070596, + "grad_norm": 1.2970024527605293, + "learning_rate": 3.4549150281252635e-06, + "loss": 0.5971, + "step": 2575 + }, + { + "epoch": 0.6123128119800333, + "grad_norm": 1.2426687443882578, + "learning_rate": 3.451253927499942e-06, + "loss": 0.5823, + "step": 2576 + }, + { + "epoch": 0.6125505110530068, + "grad_norm": 1.3305927770233337, + "learning_rate": 3.4475937451218257e-06, + "loss": 0.5545, + "step": 2577 + }, + { + "epoch": 0.6127882101259805, + "grad_norm": 1.157978933106313, + "learning_rate": 3.443934483161029e-06, + "loss": 0.5483, + "step": 2578 + }, + { + "epoch": 0.6130259091989542, + "grad_norm": 1.6842706409519277, + "learning_rate": 3.440276143787116e-06, + "loss": 0.6029, + "step": 2579 + }, + { + "epoch": 0.6132636082719277, + "grad_norm": 1.4579712610386768, + "learning_rate": 3.436618729169107e-06, + "loss": 0.6297, + "step": 2580 + }, + { + "epoch": 0.6135013073449014, + "grad_norm": 1.3241512525980823, + "learning_rate": 3.4329622414754728e-06, + "loss": 0.6186, + "step": 2581 + }, + { + "epoch": 0.6137390064178749, + "grad_norm": 1.19317364675447, + "learning_rate": 3.429306682874134e-06, + "loss": 0.5795, + "step": 2582 + }, + { + "epoch": 0.6139767054908486, + "grad_norm": 1.1726235755303547, + "learning_rate": 3.4256520555324613e-06, + "loss": 0.581, + "step": 2583 + }, + { + "epoch": 0.6142144045638221, + "grad_norm": 1.203080272814011, + "learning_rate": 3.4219983616172743e-06, + "loss": 0.5922, + "step": 2584 + }, + { + "epoch": 0.6144521036367958, + "grad_norm": 1.3734727779578555, + "learning_rate": 3.418345603294836e-06, + "loss": 0.595, + "step": 2585 + }, + { + "epoch": 0.6146898027097695, + "grad_norm": 1.1606678219115365, + "learning_rate": 3.414693782730858e-06, + "loss": 0.559, + "step": 2586 + }, + { + "epoch": 0.614927501782743, + "grad_norm": 1.3997075787983024, + "learning_rate": 3.4110429020904924e-06, + "loss": 0.5656, + "step": 2587 + }, + { + "epoch": 0.6151652008557167, + "grad_norm": 1.1141323917193737, + "learning_rate": 3.4073929635383396e-06, + "loss": 0.5622, + "step": 2588 + }, + { + "epoch": 0.6154028999286902, + "grad_norm": 2.3632489947315753, + "learning_rate": 3.4037439692384345e-06, + "loss": 0.5869, + "step": 2589 + }, + { + "epoch": 0.6156405990016639, + "grad_norm": 1.2290342944778565, + "learning_rate": 3.4000959213542573e-06, + "loss": 0.5612, + "step": 2590 + }, + { + "epoch": 0.6158782980746375, + "grad_norm": 1.226633021578343, + "learning_rate": 3.3964488220487252e-06, + "loss": 0.5546, + "step": 2591 + }, + { + "epoch": 0.6161159971476111, + "grad_norm": 1.9059136051192904, + "learning_rate": 3.3928026734841935e-06, + "loss": 0.5534, + "step": 2592 + }, + { + "epoch": 0.6163536962205848, + "grad_norm": 1.2450295802160698, + "learning_rate": 3.3891574778224524e-06, + "loss": 0.561, + "step": 2593 + }, + { + "epoch": 0.6165913952935583, + "grad_norm": 1.265635186136535, + "learning_rate": 3.385513237224729e-06, + "loss": 0.5304, + "step": 2594 + }, + { + "epoch": 0.616829094366532, + "grad_norm": 1.3232075790255364, + "learning_rate": 3.3818699538516827e-06, + "loss": 0.5585, + "step": 2595 + }, + { + "epoch": 0.6170667934395055, + "grad_norm": 1.095073094252862, + "learning_rate": 3.3782276298634076e-06, + "loss": 0.5168, + "step": 2596 + }, + { + "epoch": 0.6173044925124792, + "grad_norm": 1.2696731818382936, + "learning_rate": 3.3745862674194246e-06, + "loss": 0.5712, + "step": 2597 + }, + { + "epoch": 0.6175421915854529, + "grad_norm": 1.2983277579191106, + "learning_rate": 3.37094586867869e-06, + "loss": 0.5832, + "step": 2598 + }, + { + "epoch": 0.6177798906584264, + "grad_norm": 1.490135411163631, + "learning_rate": 3.3673064357995844e-06, + "loss": 0.5924, + "step": 2599 + }, + { + "epoch": 0.6180175897314001, + "grad_norm": 1.3354985025520811, + "learning_rate": 3.3636679709399187e-06, + "loss": 0.6295, + "step": 2600 + }, + { + "epoch": 0.6180175897314001, + "eval_loss": 0.57780522108078, + "eval_runtime": 278.2783, + "eval_samples_per_second": 39.09, + "eval_steps_per_second": 1.222, + "step": 2600 + }, + { + "epoch": 0.6182552888043736, + "grad_norm": 1.2252892307892103, + "learning_rate": 3.3600304762569267e-06, + "loss": 0.5912, + "step": 2601 + }, + { + "epoch": 0.6184929878773473, + "grad_norm": 1.1448511416149374, + "learning_rate": 3.356393953907271e-06, + "loss": 0.5743, + "step": 2602 + }, + { + "epoch": 0.6187306869503209, + "grad_norm": 1.3399066410798721, + "learning_rate": 3.352758406047033e-06, + "loss": 0.5796, + "step": 2603 + }, + { + "epoch": 0.6189683860232945, + "grad_norm": 1.1860119104713764, + "learning_rate": 3.3491238348317212e-06, + "loss": 0.5269, + "step": 2604 + }, + { + "epoch": 0.6192060850962682, + "grad_norm": 1.1318302665969515, + "learning_rate": 3.3454902424162603e-06, + "loss": 0.5653, + "step": 2605 + }, + { + "epoch": 0.6194437841692417, + "grad_norm": 1.1952962555497122, + "learning_rate": 3.341857630955e-06, + "loss": 0.6233, + "step": 2606 + }, + { + "epoch": 0.6196814832422154, + "grad_norm": 1.3890843422298458, + "learning_rate": 3.3382260026017027e-06, + "loss": 0.5815, + "step": 2607 + }, + { + "epoch": 0.6199191823151889, + "grad_norm": 1.1507451706903704, + "learning_rate": 3.3345953595095526e-06, + "loss": 0.5813, + "step": 2608 + }, + { + "epoch": 0.6201568813881626, + "grad_norm": 1.3233732061318706, + "learning_rate": 3.330965703831146e-06, + "loss": 0.5931, + "step": 2609 + }, + { + "epoch": 0.6203945804611362, + "grad_norm": 1.3729702639921961, + "learning_rate": 3.327337037718498e-06, + "loss": 0.5648, + "step": 2610 + }, + { + "epoch": 0.6206322795341098, + "grad_norm": 1.5312930911626965, + "learning_rate": 3.3237093633230323e-06, + "loss": 0.5923, + "step": 2611 + }, + { + "epoch": 0.6208699786070835, + "grad_norm": 1.3138324618351165, + "learning_rate": 3.3200826827955882e-06, + "loss": 0.5548, + "step": 2612 + }, + { + "epoch": 0.621107677680057, + "grad_norm": 1.2011062444962035, + "learning_rate": 3.3164569982864136e-06, + "loss": 0.5466, + "step": 2613 + }, + { + "epoch": 0.6213453767530307, + "grad_norm": 1.2360946081672728, + "learning_rate": 3.3128323119451654e-06, + "loss": 0.5908, + "step": 2614 + }, + { + "epoch": 0.6215830758260042, + "grad_norm": 1.209757711553855, + "learning_rate": 3.3092086259209144e-06, + "loss": 0.633, + "step": 2615 + }, + { + "epoch": 0.6218207748989779, + "grad_norm": 1.3115781322489433, + "learning_rate": 3.30558594236213e-06, + "loss": 0.5798, + "step": 2616 + }, + { + "epoch": 0.6220584739719515, + "grad_norm": 1.254709681499802, + "learning_rate": 3.301964263416693e-06, + "loss": 0.5654, + "step": 2617 + }, + { + "epoch": 0.6222961730449251, + "grad_norm": 1.7370124619642462, + "learning_rate": 3.298343591231885e-06, + "loss": 0.6006, + "step": 2618 + }, + { + "epoch": 0.6225338721178988, + "grad_norm": 1.135413767666508, + "learning_rate": 3.294723927954395e-06, + "loss": 0.5578, + "step": 2619 + }, + { + "epoch": 0.6227715711908723, + "grad_norm": 1.3363580284134202, + "learning_rate": 3.2911052757303075e-06, + "loss": 0.6072, + "step": 2620 + }, + { + "epoch": 0.623009270263846, + "grad_norm": 1.3623701773550398, + "learning_rate": 3.2874876367051154e-06, + "loss": 0.5987, + "step": 2621 + }, + { + "epoch": 0.6232469693368196, + "grad_norm": 1.1366716824569874, + "learning_rate": 3.2838710130237027e-06, + "loss": 0.564, + "step": 2622 + }, + { + "epoch": 0.6234846684097932, + "grad_norm": 1.1198841984065178, + "learning_rate": 3.2802554068303595e-06, + "loss": 0.5427, + "step": 2623 + }, + { + "epoch": 0.6237223674827668, + "grad_norm": 1.182061972905389, + "learning_rate": 3.2766408202687648e-06, + "loss": 0.5289, + "step": 2624 + }, + { + "epoch": 0.6239600665557404, + "grad_norm": 1.2523413170741045, + "learning_rate": 3.2730272554819996e-06, + "loss": 0.5628, + "step": 2625 + }, + { + "epoch": 0.6241977656287141, + "grad_norm": 1.2471316653536286, + "learning_rate": 3.269414714612534e-06, + "loss": 0.5813, + "step": 2626 + }, + { + "epoch": 0.6244354647016876, + "grad_norm": 1.4491429584064512, + "learning_rate": 3.2658031998022368e-06, + "loss": 0.6245, + "step": 2627 + }, + { + "epoch": 0.6246731637746613, + "grad_norm": 1.4700000482976152, + "learning_rate": 3.2621927131923616e-06, + "loss": 0.5936, + "step": 2628 + }, + { + "epoch": 0.6249108628476349, + "grad_norm": 1.6649809808659641, + "learning_rate": 3.2585832569235576e-06, + "loss": 0.5555, + "step": 2629 + }, + { + "epoch": 0.6251485619206085, + "grad_norm": 1.2458025226861444, + "learning_rate": 3.2549748331358593e-06, + "loss": 0.5289, + "step": 2630 + }, + { + "epoch": 0.6253862609935821, + "grad_norm": 1.3979903831404024, + "learning_rate": 3.2513674439686945e-06, + "loss": 0.6174, + "step": 2631 + }, + { + "epoch": 0.6256239600665557, + "grad_norm": 1.3881895838889327, + "learning_rate": 3.2477610915608705e-06, + "loss": 0.5821, + "step": 2632 + }, + { + "epoch": 0.6258616591395294, + "grad_norm": 1.2497218722093184, + "learning_rate": 3.244155778050586e-06, + "loss": 0.6047, + "step": 2633 + }, + { + "epoch": 0.626099358212503, + "grad_norm": 3.445386915898579, + "learning_rate": 3.24055150557542e-06, + "loss": 0.5432, + "step": 2634 + }, + { + "epoch": 0.6263370572854766, + "grad_norm": 1.2287406932787113, + "learning_rate": 3.236948276272337e-06, + "loss": 0.592, + "step": 2635 + }, + { + "epoch": 0.6265747563584502, + "grad_norm": 1.2119685730570817, + "learning_rate": 3.2333460922776794e-06, + "loss": 0.5189, + "step": 2636 + }, + { + "epoch": 0.6268124554314238, + "grad_norm": 1.1951566405222176, + "learning_rate": 3.2297449557271743e-06, + "loss": 0.5586, + "step": 2637 + }, + { + "epoch": 0.6270501545043974, + "grad_norm": 1.315108508731727, + "learning_rate": 3.226144868755923e-06, + "loss": 0.5631, + "step": 2638 + }, + { + "epoch": 0.627287853577371, + "grad_norm": 1.4297603334503395, + "learning_rate": 3.22254583349841e-06, + "loss": 0.6383, + "step": 2639 + }, + { + "epoch": 0.6275255526503447, + "grad_norm": 1.6541335242187845, + "learning_rate": 3.2189478520884897e-06, + "loss": 0.6246, + "step": 2640 + }, + { + "epoch": 0.6277632517233183, + "grad_norm": 1.3831970336030848, + "learning_rate": 3.2153509266593984e-06, + "loss": 0.5918, + "step": 2641 + }, + { + "epoch": 0.6280009507962919, + "grad_norm": 1.2105152534725871, + "learning_rate": 3.2117550593437412e-06, + "loss": 0.5165, + "step": 2642 + }, + { + "epoch": 0.6282386498692655, + "grad_norm": 1.3348860363279085, + "learning_rate": 3.2081602522734987e-06, + "loss": 0.5629, + "step": 2643 + }, + { + "epoch": 0.6284763489422391, + "grad_norm": 1.1930343974082565, + "learning_rate": 3.204566507580021e-06, + "loss": 0.595, + "step": 2644 + }, + { + "epoch": 0.6287140480152127, + "grad_norm": 1.3375961816835271, + "learning_rate": 3.2009738273940304e-06, + "loss": 0.5319, + "step": 2645 + }, + { + "epoch": 0.6289517470881864, + "grad_norm": 1.2049052484678553, + "learning_rate": 3.1973822138456157e-06, + "loss": 0.6118, + "step": 2646 + }, + { + "epoch": 0.62918944616116, + "grad_norm": 1.611177336447859, + "learning_rate": 3.1937916690642356e-06, + "loss": 0.5875, + "step": 2647 + }, + { + "epoch": 0.6294271452341336, + "grad_norm": 1.5272726768863618, + "learning_rate": 3.1902021951787125e-06, + "loss": 0.5852, + "step": 2648 + }, + { + "epoch": 0.6296648443071072, + "grad_norm": 1.2664175114021419, + "learning_rate": 3.1866137943172384e-06, + "loss": 0.5749, + "step": 2649 + }, + { + "epoch": 0.6299025433800808, + "grad_norm": 1.25654854635553, + "learning_rate": 3.183026468607362e-06, + "loss": 0.561, + "step": 2650 + }, + { + "epoch": 0.6301402424530544, + "grad_norm": 1.2491209681853255, + "learning_rate": 3.1794402201760022e-06, + "loss": 0.5757, + "step": 2651 + }, + { + "epoch": 0.630377941526028, + "grad_norm": 1.1719965233171958, + "learning_rate": 3.1758550511494336e-06, + "loss": 0.5659, + "step": 2652 + }, + { + "epoch": 0.6306156405990017, + "grad_norm": 1.0632038530787282, + "learning_rate": 3.1722709636532944e-06, + "loss": 0.5829, + "step": 2653 + }, + { + "epoch": 0.6308533396719753, + "grad_norm": 1.4415874033014402, + "learning_rate": 3.1686879598125785e-06, + "loss": 0.545, + "step": 2654 + }, + { + "epoch": 0.6310910387449489, + "grad_norm": 1.3540490109391528, + "learning_rate": 3.16510604175164e-06, + "loss": 0.5544, + "step": 2655 + }, + { + "epoch": 0.6313287378179225, + "grad_norm": 1.370797279157482, + "learning_rate": 3.161525211594187e-06, + "loss": 0.5947, + "step": 2656 + }, + { + "epoch": 0.6315664368908961, + "grad_norm": 1.4835547367791206, + "learning_rate": 3.1579454714632853e-06, + "loss": 0.5739, + "step": 2657 + }, + { + "epoch": 0.6318041359638698, + "grad_norm": 1.1688479597612824, + "learning_rate": 3.15436682348135e-06, + "loss": 0.586, + "step": 2658 + }, + { + "epoch": 0.6320418350368433, + "grad_norm": 1.2832729095301794, + "learning_rate": 3.150789269770155e-06, + "loss": 0.5518, + "step": 2659 + }, + { + "epoch": 0.632279534109817, + "grad_norm": 1.1361535221363914, + "learning_rate": 3.147212812450819e-06, + "loss": 0.5747, + "step": 2660 + }, + { + "epoch": 0.6325172331827906, + "grad_norm": 1.115575433443054, + "learning_rate": 3.1436374536438154e-06, + "loss": 0.5724, + "step": 2661 + }, + { + "epoch": 0.6327549322557642, + "grad_norm": 1.2559017464958635, + "learning_rate": 3.1400631954689626e-06, + "loss": 0.5413, + "step": 2662 + }, + { + "epoch": 0.6329926313287378, + "grad_norm": 1.216407790482303, + "learning_rate": 3.13649004004543e-06, + "loss": 0.5621, + "step": 2663 + }, + { + "epoch": 0.6332303304017114, + "grad_norm": 1.4128531192306077, + "learning_rate": 3.1329179894917306e-06, + "loss": 0.5554, + "step": 2664 + }, + { + "epoch": 0.6334680294746851, + "grad_norm": 1.33476867054213, + "learning_rate": 3.1293470459257237e-06, + "loss": 0.558, + "step": 2665 + }, + { + "epoch": 0.6337057285476586, + "grad_norm": 1.3222316635927374, + "learning_rate": 3.12577721146461e-06, + "loss": 0.559, + "step": 2666 + }, + { + "epoch": 0.6339434276206323, + "grad_norm": 1.2041037510467123, + "learning_rate": 3.1222084882249375e-06, + "loss": 0.5153, + "step": 2667 + }, + { + "epoch": 0.6341811266936059, + "grad_norm": 1.5666125465480047, + "learning_rate": 3.118640878322589e-06, + "loss": 0.5404, + "step": 2668 + }, + { + "epoch": 0.6344188257665795, + "grad_norm": 1.352528421530246, + "learning_rate": 3.1150743838727925e-06, + "loss": 0.5665, + "step": 2669 + }, + { + "epoch": 0.6346565248395531, + "grad_norm": 1.2436744257897168, + "learning_rate": 3.1115090069901117e-06, + "loss": 0.5856, + "step": 2670 + }, + { + "epoch": 0.6348942239125267, + "grad_norm": 1.6023050948760091, + "learning_rate": 3.107944749788449e-06, + "loss": 0.6004, + "step": 2671 + }, + { + "epoch": 0.6351319229855004, + "grad_norm": 1.3559375955249808, + "learning_rate": 3.104381614381041e-06, + "loss": 0.5703, + "step": 2672 + }, + { + "epoch": 0.635369622058474, + "grad_norm": 1.3409679769058098, + "learning_rate": 3.100819602880464e-06, + "loss": 0.5914, + "step": 2673 + }, + { + "epoch": 0.6356073211314476, + "grad_norm": 1.48430305774554, + "learning_rate": 3.0972587173986214e-06, + "loss": 0.5533, + "step": 2674 + }, + { + "epoch": 0.6358450202044212, + "grad_norm": 1.105166188195845, + "learning_rate": 3.093698960046754e-06, + "loss": 0.5218, + "step": 2675 + }, + { + "epoch": 0.6360827192773948, + "grad_norm": 1.259609587573441, + "learning_rate": 3.0901403329354298e-06, + "loss": 0.5294, + "step": 2676 + }, + { + "epoch": 0.6363204183503685, + "grad_norm": 1.4467630766670165, + "learning_rate": 3.0865828381745515e-06, + "loss": 0.5227, + "step": 2677 + }, + { + "epoch": 0.636558117423342, + "grad_norm": 1.2197557720018555, + "learning_rate": 3.0830264778733475e-06, + "loss": 0.5752, + "step": 2678 + }, + { + "epoch": 0.6367958164963157, + "grad_norm": 1.1993013683257816, + "learning_rate": 3.0794712541403716e-06, + "loss": 0.5276, + "step": 2679 + }, + { + "epoch": 0.6370335155692893, + "grad_norm": 1.2299287878184768, + "learning_rate": 3.075917169083508e-06, + "loss": 0.5625, + "step": 2680 + }, + { + "epoch": 0.6372712146422629, + "grad_norm": 1.3092438073968722, + "learning_rate": 3.072364224809963e-06, + "loss": 0.5885, + "step": 2681 + }, + { + "epoch": 0.6375089137152365, + "grad_norm": 1.293687785832519, + "learning_rate": 3.0688124234262677e-06, + "loss": 0.6239, + "step": 2682 + }, + { + "epoch": 0.6377466127882101, + "grad_norm": 1.277006085733343, + "learning_rate": 3.0652617670382745e-06, + "loss": 0.5779, + "step": 2683 + }, + { + "epoch": 0.6379843118611838, + "grad_norm": 1.10594079419004, + "learning_rate": 3.0617122577511587e-06, + "loss": 0.5918, + "step": 2684 + }, + { + "epoch": 0.6382220109341573, + "grad_norm": 1.4199277773042267, + "learning_rate": 3.0581638976694127e-06, + "loss": 0.5864, + "step": 2685 + }, + { + "epoch": 0.638459710007131, + "grad_norm": 1.4301190134689887, + "learning_rate": 3.054616688896852e-06, + "loss": 0.6007, + "step": 2686 + }, + { + "epoch": 0.6386974090801046, + "grad_norm": 1.706380600538607, + "learning_rate": 3.0510706335366034e-06, + "loss": 0.5602, + "step": 2687 + }, + { + "epoch": 0.6389351081530782, + "grad_norm": 1.2803883164638665, + "learning_rate": 3.0475257336911167e-06, + "loss": 0.5489, + "step": 2688 + }, + { + "epoch": 0.6391728072260519, + "grad_norm": 1.307128354360484, + "learning_rate": 3.04398199146215e-06, + "loss": 0.62, + "step": 2689 + }, + { + "epoch": 0.6394105062990254, + "grad_norm": 1.2413197256488055, + "learning_rate": 3.0404394089507805e-06, + "loss": 0.5718, + "step": 2690 + }, + { + "epoch": 0.6396482053719991, + "grad_norm": 1.200256235145322, + "learning_rate": 3.0368979882573926e-06, + "loss": 0.5583, + "step": 2691 + }, + { + "epoch": 0.6398859044449726, + "grad_norm": 1.2671629862139187, + "learning_rate": 3.0333577314816883e-06, + "loss": 0.565, + "step": 2692 + }, + { + "epoch": 0.6401236035179463, + "grad_norm": 1.199123417915596, + "learning_rate": 3.0298186407226725e-06, + "loss": 0.5472, + "step": 2693 + }, + { + "epoch": 0.64036130259092, + "grad_norm": 2.229623207861314, + "learning_rate": 3.0262807180786647e-06, + "loss": 0.5716, + "step": 2694 + }, + { + "epoch": 0.6405990016638935, + "grad_norm": 1.3613844959740735, + "learning_rate": 3.0227439656472878e-06, + "loss": 0.5161, + "step": 2695 + }, + { + "epoch": 0.6408367007368672, + "grad_norm": 1.2370916744814928, + "learning_rate": 3.019208385525474e-06, + "loss": 0.6189, + "step": 2696 + }, + { + "epoch": 0.6410743998098407, + "grad_norm": 1.099744581693413, + "learning_rate": 3.015673979809457e-06, + "loss": 0.5148, + "step": 2697 + }, + { + "epoch": 0.6413120988828144, + "grad_norm": 1.20758541550779, + "learning_rate": 3.0121407505947774e-06, + "loss": 0.5679, + "step": 2698 + }, + { + "epoch": 0.6415497979557879, + "grad_norm": 1.2095376036954955, + "learning_rate": 3.0086086999762755e-06, + "loss": 0.5377, + "step": 2699 + }, + { + "epoch": 0.6417874970287616, + "grad_norm": 1.9141021627784485, + "learning_rate": 3.0050778300480963e-06, + "loss": 0.556, + "step": 2700 + }, + { + "epoch": 0.6417874970287616, + "eval_loss": 0.5743858218193054, + "eval_runtime": 278.4853, + "eval_samples_per_second": 39.061, + "eval_steps_per_second": 1.221, + "step": 2700 + }, + { + "epoch": 0.6420251961017353, + "grad_norm": 1.273922844380647, + "learning_rate": 3.0015481429036807e-06, + "loss": 0.5921, + "step": 2701 + }, + { + "epoch": 0.6422628951747088, + "grad_norm": 1.4659329782499264, + "learning_rate": 2.998019640635772e-06, + "loss": 0.5689, + "step": 2702 + }, + { + "epoch": 0.6425005942476825, + "grad_norm": 1.097249165796798, + "learning_rate": 2.9944923253364066e-06, + "loss": 0.5802, + "step": 2703 + }, + { + "epoch": 0.642738293320656, + "grad_norm": 1.3502087475359599, + "learning_rate": 2.990966199096924e-06, + "loss": 0.5676, + "step": 2704 + }, + { + "epoch": 0.6429759923936297, + "grad_norm": 1.292302284907063, + "learning_rate": 2.9874412640079504e-06, + "loss": 0.5273, + "step": 2705 + }, + { + "epoch": 0.6432136914666032, + "grad_norm": 1.0867462216438901, + "learning_rate": 2.983917522159413e-06, + "loss": 0.5122, + "step": 2706 + }, + { + "epoch": 0.6434513905395769, + "grad_norm": 1.2884065385557248, + "learning_rate": 2.980394975640526e-06, + "loss": 0.6064, + "step": 2707 + }, + { + "epoch": 0.6436890896125506, + "grad_norm": 1.4657860842289085, + "learning_rate": 2.9768736265398e-06, + "loss": 0.6036, + "step": 2708 + }, + { + "epoch": 0.6439267886855241, + "grad_norm": 1.0661982626000106, + "learning_rate": 2.97335347694503e-06, + "loss": 0.5435, + "step": 2709 + }, + { + "epoch": 0.6441644877584978, + "grad_norm": 1.316458618352894, + "learning_rate": 2.969834528943306e-06, + "loss": 0.5818, + "step": 2710 + }, + { + "epoch": 0.6444021868314713, + "grad_norm": 1.4227334687638653, + "learning_rate": 2.966316784621e-06, + "loss": 0.578, + "step": 2711 + }, + { + "epoch": 0.644639885904445, + "grad_norm": 1.2913760309852513, + "learning_rate": 2.962800246063774e-06, + "loss": 0.5883, + "step": 2712 + }, + { + "epoch": 0.6448775849774185, + "grad_norm": 1.2540891656986666, + "learning_rate": 2.9592849153565727e-06, + "loss": 0.556, + "step": 2713 + }, + { + "epoch": 0.6451152840503922, + "grad_norm": 1.3694133674683955, + "learning_rate": 2.955770794583628e-06, + "loss": 0.5349, + "step": 2714 + }, + { + "epoch": 0.6453529831233659, + "grad_norm": 1.2551272239071538, + "learning_rate": 2.9522578858284496e-06, + "loss": 0.5479, + "step": 2715 + }, + { + "epoch": 0.6455906821963394, + "grad_norm": 1.26280488608376, + "learning_rate": 2.9487461911738334e-06, + "loss": 0.542, + "step": 2716 + }, + { + "epoch": 0.6458283812693131, + "grad_norm": 1.4763257007285062, + "learning_rate": 2.9452357127018516e-06, + "loss": 0.6036, + "step": 2717 + }, + { + "epoch": 0.6460660803422866, + "grad_norm": 1.282966449181743, + "learning_rate": 2.9417264524938587e-06, + "loss": 0.6026, + "step": 2718 + }, + { + "epoch": 0.6463037794152603, + "grad_norm": 1.3077761876734846, + "learning_rate": 2.9382184126304834e-06, + "loss": 0.538, + "step": 2719 + }, + { + "epoch": 0.6465414784882338, + "grad_norm": 1.1992987267796003, + "learning_rate": 2.934711595191635e-06, + "loss": 0.5314, + "step": 2720 + }, + { + "epoch": 0.6467791775612075, + "grad_norm": 1.4247297870753877, + "learning_rate": 2.931206002256493e-06, + "loss": 0.6074, + "step": 2721 + }, + { + "epoch": 0.6470168766341812, + "grad_norm": 1.3061745760408268, + "learning_rate": 2.9277016359035165e-06, + "loss": 0.5306, + "step": 2722 + }, + { + "epoch": 0.6472545757071547, + "grad_norm": 1.2321697241664522, + "learning_rate": 2.924198498210432e-06, + "loss": 0.543, + "step": 2723 + }, + { + "epoch": 0.6474922747801284, + "grad_norm": 1.2758610203910912, + "learning_rate": 2.920696591254242e-06, + "loss": 0.6556, + "step": 2724 + }, + { + "epoch": 0.6477299738531019, + "grad_norm": 1.1689930819124865, + "learning_rate": 2.917195917111215e-06, + "loss": 0.5668, + "step": 2725 + }, + { + "epoch": 0.6479676729260756, + "grad_norm": 1.2398361367343929, + "learning_rate": 2.9136964778568932e-06, + "loss": 0.6169, + "step": 2726 + }, + { + "epoch": 0.6482053719990492, + "grad_norm": 1.2668309504428064, + "learning_rate": 2.910198275566085e-06, + "loss": 0.5643, + "step": 2727 + }, + { + "epoch": 0.6484430710720228, + "grad_norm": 1.127410109200658, + "learning_rate": 2.906701312312861e-06, + "loss": 0.5926, + "step": 2728 + }, + { + "epoch": 0.6486807701449965, + "grad_norm": 1.2871041831888101, + "learning_rate": 2.9032055901705634e-06, + "loss": 0.5872, + "step": 2729 + }, + { + "epoch": 0.64891846921797, + "grad_norm": 1.243940222888916, + "learning_rate": 2.899711111211796e-06, + "loss": 0.578, + "step": 2730 + }, + { + "epoch": 0.6491561682909437, + "grad_norm": 1.121119340184251, + "learning_rate": 2.8962178775084267e-06, + "loss": 0.5365, + "step": 2731 + }, + { + "epoch": 0.6493938673639172, + "grad_norm": 1.2475179642752026, + "learning_rate": 2.892725891131581e-06, + "loss": 0.531, + "step": 2732 + }, + { + "epoch": 0.6496315664368909, + "grad_norm": 1.4695313231887472, + "learning_rate": 2.889235154151649e-06, + "loss": 0.6105, + "step": 2733 + }, + { + "epoch": 0.6498692655098645, + "grad_norm": 1.3641831049416935, + "learning_rate": 2.8857456686382794e-06, + "loss": 0.545, + "step": 2734 + }, + { + "epoch": 0.6501069645828381, + "grad_norm": 1.5217167767351896, + "learning_rate": 2.8822574366603804e-06, + "loss": 0.5912, + "step": 2735 + }, + { + "epoch": 0.6503446636558118, + "grad_norm": 1.5346827231406175, + "learning_rate": 2.878770460286111e-06, + "loss": 0.5699, + "step": 2736 + }, + { + "epoch": 0.6505823627287853, + "grad_norm": 1.469622834837515, + "learning_rate": 2.8752847415828923e-06, + "loss": 0.5426, + "step": 2737 + }, + { + "epoch": 0.650820061801759, + "grad_norm": 1.083627871331437, + "learning_rate": 2.8718002826173953e-06, + "loss": 0.5178, + "step": 2738 + }, + { + "epoch": 0.6510577608747325, + "grad_norm": 1.1181645434774197, + "learning_rate": 2.8683170854555497e-06, + "loss": 0.5058, + "step": 2739 + }, + { + "epoch": 0.6512954599477062, + "grad_norm": 1.1328068190154184, + "learning_rate": 2.8648351521625284e-06, + "loss": 0.536, + "step": 2740 + }, + { + "epoch": 0.6515331590206798, + "grad_norm": 1.1831519225775853, + "learning_rate": 2.86135448480276e-06, + "loss": 0.6028, + "step": 2741 + }, + { + "epoch": 0.6517708580936534, + "grad_norm": 1.2436829876105273, + "learning_rate": 2.8578750854399294e-06, + "loss": 0.5614, + "step": 2742 + }, + { + "epoch": 0.6520085571666271, + "grad_norm": 1.1934752673743247, + "learning_rate": 2.8543969561369556e-06, + "loss": 0.5495, + "step": 2743 + }, + { + "epoch": 0.6522462562396006, + "grad_norm": 1.5476276666162068, + "learning_rate": 2.850920098956014e-06, + "loss": 0.5989, + "step": 2744 + }, + { + "epoch": 0.6524839553125743, + "grad_norm": 3.0002838272934893, + "learning_rate": 2.8474445159585235e-06, + "loss": 0.5756, + "step": 2745 + }, + { + "epoch": 0.6527216543855479, + "grad_norm": 1.724287262066002, + "learning_rate": 2.8439702092051497e-06, + "loss": 0.5979, + "step": 2746 + }, + { + "epoch": 0.6529593534585215, + "grad_norm": 1.278851224249109, + "learning_rate": 2.8404971807557957e-06, + "loss": 0.5771, + "step": 2747 + }, + { + "epoch": 0.6531970525314952, + "grad_norm": 1.1713718450426323, + "learning_rate": 2.8370254326696113e-06, + "loss": 0.5465, + "step": 2748 + }, + { + "epoch": 0.6534347516044687, + "grad_norm": 1.343102255520692, + "learning_rate": 2.8335549670049866e-06, + "loss": 0.5373, + "step": 2749 + }, + { + "epoch": 0.6536724506774424, + "grad_norm": 1.151431663735979, + "learning_rate": 2.8300857858195533e-06, + "loss": 0.5636, + "step": 2750 + }, + { + "epoch": 0.653910149750416, + "grad_norm": 1.1863737013774638, + "learning_rate": 2.8266178911701757e-06, + "loss": 0.5526, + "step": 2751 + }, + { + "epoch": 0.6541478488233896, + "grad_norm": 1.19015430332238, + "learning_rate": 2.8231512851129596e-06, + "loss": 0.5787, + "step": 2752 + }, + { + "epoch": 0.6543855478963632, + "grad_norm": 1.4919069048855993, + "learning_rate": 2.8196859697032463e-06, + "loss": 0.539, + "step": 2753 + }, + { + "epoch": 0.6546232469693368, + "grad_norm": 1.1080057760054194, + "learning_rate": 2.8162219469956144e-06, + "loss": 0.5525, + "step": 2754 + }, + { + "epoch": 0.6548609460423105, + "grad_norm": 1.290615564919263, + "learning_rate": 2.812759219043869e-06, + "loss": 0.5446, + "step": 2755 + }, + { + "epoch": 0.655098645115284, + "grad_norm": 1.5199782532164012, + "learning_rate": 2.809297787901053e-06, + "loss": 0.5594, + "step": 2756 + }, + { + "epoch": 0.6553363441882577, + "grad_norm": 1.3811127467237063, + "learning_rate": 2.80583765561944e-06, + "loss": 0.5476, + "step": 2757 + }, + { + "epoch": 0.6555740432612313, + "grad_norm": 1.2187653919345753, + "learning_rate": 2.8023788242505345e-06, + "loss": 0.5588, + "step": 2758 + }, + { + "epoch": 0.6558117423342049, + "grad_norm": 1.479861583174111, + "learning_rate": 2.798921295845064e-06, + "loss": 0.596, + "step": 2759 + }, + { + "epoch": 0.6560494414071785, + "grad_norm": 1.3938099934772876, + "learning_rate": 2.7954650724529887e-06, + "loss": 0.5535, + "step": 2760 + }, + { + "epoch": 0.6562871404801521, + "grad_norm": 1.3939101052432854, + "learning_rate": 2.7920101561234954e-06, + "loss": 0.5423, + "step": 2761 + }, + { + "epoch": 0.6565248395531258, + "grad_norm": 1.2431753758914272, + "learning_rate": 2.7885565489049948e-06, + "loss": 0.5291, + "step": 2762 + }, + { + "epoch": 0.6567625386260993, + "grad_norm": 1.3107979569522534, + "learning_rate": 2.7851042528451174e-06, + "loss": 0.582, + "step": 2763 + }, + { + "epoch": 0.657000237699073, + "grad_norm": 1.2793391488621009, + "learning_rate": 2.7816532699907217e-06, + "loss": 0.5185, + "step": 2764 + }, + { + "epoch": 0.6572379367720466, + "grad_norm": 1.2169916982038105, + "learning_rate": 2.778203602387886e-06, + "loss": 0.5228, + "step": 2765 + }, + { + "epoch": 0.6574756358450202, + "grad_norm": 1.8104586737133965, + "learning_rate": 2.774755252081911e-06, + "loss": 0.523, + "step": 2766 + }, + { + "epoch": 0.6577133349179938, + "grad_norm": 1.4210284546216887, + "learning_rate": 2.771308221117309e-06, + "loss": 0.5652, + "step": 2767 + }, + { + "epoch": 0.6579510339909674, + "grad_norm": 1.2564662720909836, + "learning_rate": 2.767862511537818e-06, + "loss": 0.5134, + "step": 2768 + }, + { + "epoch": 0.6581887330639411, + "grad_norm": 1.5000749235125677, + "learning_rate": 2.7644181253863882e-06, + "loss": 0.6062, + "step": 2769 + }, + { + "epoch": 0.6584264321369147, + "grad_norm": 1.3265209547285506, + "learning_rate": 2.76097506470519e-06, + "loss": 0.5654, + "step": 2770 + }, + { + "epoch": 0.6586641312098883, + "grad_norm": 1.3438034667920224, + "learning_rate": 2.7575333315355994e-06, + "loss": 0.595, + "step": 2771 + }, + { + "epoch": 0.6589018302828619, + "grad_norm": 1.3354167024078478, + "learning_rate": 2.754092927918213e-06, + "loss": 0.617, + "step": 2772 + }, + { + "epoch": 0.6591395293558355, + "grad_norm": 1.347640271829174, + "learning_rate": 2.750653855892836e-06, + "loss": 0.5635, + "step": 2773 + }, + { + "epoch": 0.6593772284288091, + "grad_norm": 1.3064703923884364, + "learning_rate": 2.747216117498487e-06, + "loss": 0.5713, + "step": 2774 + }, + { + "epoch": 0.6596149275017827, + "grad_norm": 1.1320663327683442, + "learning_rate": 2.743779714773386e-06, + "loss": 0.515, + "step": 2775 + }, + { + "epoch": 0.6598526265747564, + "grad_norm": 1.2317911035986544, + "learning_rate": 2.7403446497549714e-06, + "loss": 0.537, + "step": 2776 + }, + { + "epoch": 0.66009032564773, + "grad_norm": 1.2691292311074556, + "learning_rate": 2.736910924479881e-06, + "loss": 0.5558, + "step": 2777 + }, + { + "epoch": 0.6603280247207036, + "grad_norm": 1.1373342999404463, + "learning_rate": 2.7334785409839637e-06, + "loss": 0.5929, + "step": 2778 + }, + { + "epoch": 0.6605657237936772, + "grad_norm": 1.2415329353801317, + "learning_rate": 2.7300475013022666e-06, + "loss": 0.5867, + "step": 2779 + }, + { + "epoch": 0.6608034228666508, + "grad_norm": 1.4892453402920556, + "learning_rate": 2.726617807469045e-06, + "loss": 0.5828, + "step": 2780 + }, + { + "epoch": 0.6610411219396244, + "grad_norm": 1.343109558306315, + "learning_rate": 2.7231894615177546e-06, + "loss": 0.5576, + "step": 2781 + }, + { + "epoch": 0.661278821012598, + "grad_norm": 1.2332693592536144, + "learning_rate": 2.719762465481055e-06, + "loss": 0.5908, + "step": 2782 + }, + { + "epoch": 0.6615165200855717, + "grad_norm": 1.3253753639723485, + "learning_rate": 2.7163368213907975e-06, + "loss": 0.5524, + "step": 2783 + }, + { + "epoch": 0.6617542191585453, + "grad_norm": 1.1939055970098575, + "learning_rate": 2.7129125312780396e-06, + "loss": 0.5343, + "step": 2784 + }, + { + "epoch": 0.6619919182315189, + "grad_norm": 1.1919785696894032, + "learning_rate": 2.7094895971730326e-06, + "loss": 0.5458, + "step": 2785 + }, + { + "epoch": 0.6622296173044925, + "grad_norm": 1.2488312116759763, + "learning_rate": 2.706068021105227e-06, + "loss": 0.6073, + "step": 2786 + }, + { + "epoch": 0.6624673163774661, + "grad_norm": 1.180623018671934, + "learning_rate": 2.7026478051032625e-06, + "loss": 0.5407, + "step": 2787 + }, + { + "epoch": 0.6627050154504397, + "grad_norm": 1.1710583191769721, + "learning_rate": 2.699228951194976e-06, + "loss": 0.5344, + "step": 2788 + }, + { + "epoch": 0.6629427145234134, + "grad_norm": 1.417223394804396, + "learning_rate": 2.695811461407397e-06, + "loss": 0.5016, + "step": 2789 + }, + { + "epoch": 0.663180413596387, + "grad_norm": 1.1678890279308698, + "learning_rate": 2.692395337766748e-06, + "loss": 0.5865, + "step": 2790 + }, + { + "epoch": 0.6634181126693606, + "grad_norm": 1.3010903060825139, + "learning_rate": 2.6889805822984348e-06, + "loss": 0.5684, + "step": 2791 + }, + { + "epoch": 0.6636558117423342, + "grad_norm": 1.311641124820533, + "learning_rate": 2.685567197027058e-06, + "loss": 0.5641, + "step": 2792 + }, + { + "epoch": 0.6638935108153078, + "grad_norm": 1.195718936438215, + "learning_rate": 2.6821551839764055e-06, + "loss": 0.6088, + "step": 2793 + }, + { + "epoch": 0.6641312098882814, + "grad_norm": 1.065530889789563, + "learning_rate": 2.6787445451694506e-06, + "loss": 0.4797, + "step": 2794 + }, + { + "epoch": 0.664368908961255, + "grad_norm": 1.2516178728753595, + "learning_rate": 2.6753352826283484e-06, + "loss": 0.5638, + "step": 2795 + }, + { + "epoch": 0.6646066080342287, + "grad_norm": 1.3728870971201688, + "learning_rate": 2.671927398374443e-06, + "loss": 0.5518, + "step": 2796 + }, + { + "epoch": 0.6648443071072023, + "grad_norm": 1.1729306550069665, + "learning_rate": 2.668520894428259e-06, + "loss": 0.5777, + "step": 2797 + }, + { + "epoch": 0.6650820061801759, + "grad_norm": 1.3861154006377854, + "learning_rate": 2.6651157728095066e-06, + "loss": 0.5409, + "step": 2798 + }, + { + "epoch": 0.6653197052531495, + "grad_norm": 1.1140011403331467, + "learning_rate": 2.6617120355370667e-06, + "loss": 0.5664, + "step": 2799 + }, + { + "epoch": 0.6655574043261231, + "grad_norm": 1.2597475372974463, + "learning_rate": 2.658309684629009e-06, + "loss": 0.5257, + "step": 2800 + }, + { + "epoch": 0.6655574043261231, + "eval_loss": 0.5708922743797302, + "eval_runtime": 278.3806, + "eval_samples_per_second": 39.076, + "eval_steps_per_second": 1.221, + "step": 2800 + }, + { + "epoch": 0.6657951033990968, + "grad_norm": 1.3757051135934466, + "learning_rate": 2.6549087221025772e-06, + "loss": 0.5882, + "step": 2801 + }, + { + "epoch": 0.6660328024720703, + "grad_norm": 1.6125132364930466, + "learning_rate": 2.6515091499741946e-06, + "loss": 0.5374, + "step": 2802 + }, + { + "epoch": 0.666270501545044, + "grad_norm": 1.4327239708618422, + "learning_rate": 2.648110970259454e-06, + "loss": 0.558, + "step": 2803 + }, + { + "epoch": 0.6665082006180176, + "grad_norm": 1.2643468375614613, + "learning_rate": 2.6447141849731283e-06, + "loss": 0.6213, + "step": 2804 + }, + { + "epoch": 0.6667458996909912, + "grad_norm": 1.2548479743344376, + "learning_rate": 2.6413187961291632e-06, + "loss": 0.537, + "step": 2805 + }, + { + "epoch": 0.6669835987639648, + "grad_norm": 1.1692344963164203, + "learning_rate": 2.637924805740676e-06, + "loss": 0.5509, + "step": 2806 + }, + { + "epoch": 0.6672212978369384, + "grad_norm": 1.1788162184963673, + "learning_rate": 2.6345322158199503e-06, + "loss": 0.5467, + "step": 2807 + }, + { + "epoch": 0.6674589969099121, + "grad_norm": 1.3856432684583442, + "learning_rate": 2.6311410283784457e-06, + "loss": 0.5793, + "step": 2808 + }, + { + "epoch": 0.6676966959828856, + "grad_norm": 1.141543563020898, + "learning_rate": 2.6277512454267874e-06, + "loss": 0.5915, + "step": 2809 + }, + { + "epoch": 0.6679343950558593, + "grad_norm": 1.195390164548544, + "learning_rate": 2.6243628689747695e-06, + "loss": 0.5768, + "step": 2810 + }, + { + "epoch": 0.6681720941288329, + "grad_norm": 1.316705497019411, + "learning_rate": 2.620975901031348e-06, + "loss": 0.5968, + "step": 2811 + }, + { + "epoch": 0.6684097932018065, + "grad_norm": 1.4535854592571031, + "learning_rate": 2.617590343604648e-06, + "loss": 0.5519, + "step": 2812 + }, + { + "epoch": 0.6686474922747802, + "grad_norm": 1.5488712169089098, + "learning_rate": 2.614206198701958e-06, + "loss": 0.6027, + "step": 2813 + }, + { + "epoch": 0.6688851913477537, + "grad_norm": 1.379607345921298, + "learning_rate": 2.6108234683297284e-06, + "loss": 0.5219, + "step": 2814 + }, + { + "epoch": 0.6691228904207274, + "grad_norm": 1.2351483211836027, + "learning_rate": 2.607442154493568e-06, + "loss": 0.5708, + "step": 2815 + }, + { + "epoch": 0.6693605894937009, + "grad_norm": 1.448161586824617, + "learning_rate": 2.60406225919825e-06, + "loss": 0.5451, + "step": 2816 + }, + { + "epoch": 0.6695982885666746, + "grad_norm": 1.5398318102412345, + "learning_rate": 2.600683784447704e-06, + "loss": 0.5939, + "step": 2817 + }, + { + "epoch": 0.6698359876396482, + "grad_norm": 1.2380300560376145, + "learning_rate": 2.597306732245021e-06, + "loss": 0.5348, + "step": 2818 + }, + { + "epoch": 0.6700736867126218, + "grad_norm": 1.45617243369074, + "learning_rate": 2.593931104592442e-06, + "loss": 0.5959, + "step": 2819 + }, + { + "epoch": 0.6703113857855955, + "grad_norm": 1.1770287273032345, + "learning_rate": 2.590556903491369e-06, + "loss": 0.5543, + "step": 2820 + }, + { + "epoch": 0.670549084858569, + "grad_norm": 1.3675486992554498, + "learning_rate": 2.5871841309423557e-06, + "loss": 0.5321, + "step": 2821 + }, + { + "epoch": 0.6707867839315427, + "grad_norm": 1.2056669263723936, + "learning_rate": 2.5838127889451113e-06, + "loss": 0.5921, + "step": 2822 + }, + { + "epoch": 0.6710244830045163, + "grad_norm": 1.1339763461961614, + "learning_rate": 2.5804428794984926e-06, + "loss": 0.5435, + "step": 2823 + }, + { + "epoch": 0.6712621820774899, + "grad_norm": 1.4570429303384902, + "learning_rate": 2.577074404600511e-06, + "loss": 0.562, + "step": 2824 + }, + { + "epoch": 0.6714998811504636, + "grad_norm": 1.223172825498899, + "learning_rate": 2.573707366248325e-06, + "loss": 0.5973, + "step": 2825 + }, + { + "epoch": 0.6717375802234371, + "grad_norm": 1.3255867175562517, + "learning_rate": 2.5703417664382437e-06, + "loss": 0.5192, + "step": 2826 + }, + { + "epoch": 0.6719752792964108, + "grad_norm": 1.2027380043915763, + "learning_rate": 2.5669776071657194e-06, + "loss": 0.5397, + "step": 2827 + }, + { + "epoch": 0.6722129783693843, + "grad_norm": 1.3786350487834393, + "learning_rate": 2.5636148904253545e-06, + "loss": 0.553, + "step": 2828 + }, + { + "epoch": 0.672450677442358, + "grad_norm": 1.2223144525754226, + "learning_rate": 2.560253618210893e-06, + "loss": 0.6006, + "step": 2829 + }, + { + "epoch": 0.6726883765153316, + "grad_norm": 1.3635040890920542, + "learning_rate": 2.5568937925152272e-06, + "loss": 0.5573, + "step": 2830 + }, + { + "epoch": 0.6729260755883052, + "grad_norm": 1.2919763411969063, + "learning_rate": 2.5535354153303827e-06, + "loss": 0.5727, + "step": 2831 + }, + { + "epoch": 0.6731637746612789, + "grad_norm": 1.73454755960776, + "learning_rate": 2.550178488647536e-06, + "loss": 0.5607, + "step": 2832 + }, + { + "epoch": 0.6734014737342524, + "grad_norm": 1.273312923112473, + "learning_rate": 2.546823014456998e-06, + "loss": 0.5641, + "step": 2833 + }, + { + "epoch": 0.6736391728072261, + "grad_norm": 1.3033361586502223, + "learning_rate": 2.5434689947482226e-06, + "loss": 0.5377, + "step": 2834 + }, + { + "epoch": 0.6738768718801996, + "grad_norm": 1.5932622874751665, + "learning_rate": 2.540116431509795e-06, + "loss": 0.5708, + "step": 2835 + }, + { + "epoch": 0.6741145709531733, + "grad_norm": 1.4434566024671653, + "learning_rate": 2.5367653267294413e-06, + "loss": 0.5758, + "step": 2836 + }, + { + "epoch": 0.674352270026147, + "grad_norm": 1.1483050049610763, + "learning_rate": 2.5334156823940237e-06, + "loss": 0.5519, + "step": 2837 + }, + { + "epoch": 0.6745899690991205, + "grad_norm": 1.1209273963222586, + "learning_rate": 2.530067500489537e-06, + "loss": 0.5397, + "step": 2838 + }, + { + "epoch": 0.6748276681720942, + "grad_norm": 1.1732604431826807, + "learning_rate": 2.526720783001107e-06, + "loss": 0.5267, + "step": 2839 + }, + { + "epoch": 0.6750653672450677, + "grad_norm": 1.357616619477925, + "learning_rate": 2.5233755319129936e-06, + "loss": 0.5966, + "step": 2840 + }, + { + "epoch": 0.6753030663180414, + "grad_norm": 1.2071940019428913, + "learning_rate": 2.520031749208588e-06, + "loss": 0.5803, + "step": 2841 + }, + { + "epoch": 0.6755407653910149, + "grad_norm": 1.2528950730313027, + "learning_rate": 2.51668943687041e-06, + "loss": 0.5907, + "step": 2842 + }, + { + "epoch": 0.6757784644639886, + "grad_norm": 1.079825590772035, + "learning_rate": 2.513348596880105e-06, + "loss": 0.5435, + "step": 2843 + }, + { + "epoch": 0.6760161635369623, + "grad_norm": 1.176384937483182, + "learning_rate": 2.5100092312184483e-06, + "loss": 0.5352, + "step": 2844 + }, + { + "epoch": 0.6762538626099358, + "grad_norm": 1.3488903985299727, + "learning_rate": 2.506671341865341e-06, + "loss": 0.5709, + "step": 2845 + }, + { + "epoch": 0.6764915616829095, + "grad_norm": 1.3800507229018417, + "learning_rate": 2.50333493079981e-06, + "loss": 0.5676, + "step": 2846 + }, + { + "epoch": 0.676729260755883, + "grad_norm": 2.068695140517832, + "learning_rate": 2.5000000000000015e-06, + "loss": 0.5821, + "step": 2847 + }, + { + "epoch": 0.6769669598288567, + "grad_norm": 1.1452231099677486, + "learning_rate": 2.4966665514431863e-06, + "loss": 0.506, + "step": 2848 + }, + { + "epoch": 0.6772046589018302, + "grad_norm": 1.2745610708502004, + "learning_rate": 2.4933345871057585e-06, + "loss": 0.5732, + "step": 2849 + }, + { + "epoch": 0.6774423579748039, + "grad_norm": 1.2424118454802393, + "learning_rate": 2.4900041089632314e-06, + "loss": 0.5956, + "step": 2850 + }, + { + "epoch": 0.6776800570477776, + "grad_norm": 1.4154305182437286, + "learning_rate": 2.486675118990233e-06, + "loss": 0.5795, + "step": 2851 + }, + { + "epoch": 0.6779177561207511, + "grad_norm": 1.2353682686617211, + "learning_rate": 2.4833476191605136e-06, + "loss": 0.5677, + "step": 2852 + }, + { + "epoch": 0.6781554551937248, + "grad_norm": 1.2015892978955212, + "learning_rate": 2.4800216114469388e-06, + "loss": 0.6003, + "step": 2853 + }, + { + "epoch": 0.6783931542666983, + "grad_norm": 1.153419405690255, + "learning_rate": 2.476697097821491e-06, + "loss": 0.5354, + "step": 2854 + }, + { + "epoch": 0.678630853339672, + "grad_norm": 1.2059845917717982, + "learning_rate": 2.473374080255261e-06, + "loss": 0.55, + "step": 2855 + }, + { + "epoch": 0.6788685524126455, + "grad_norm": 1.150327473627533, + "learning_rate": 2.470052560718459e-06, + "loss": 0.5408, + "step": 2856 + }, + { + "epoch": 0.6791062514856192, + "grad_norm": 1.2687748986674892, + "learning_rate": 2.466732541180404e-06, + "loss": 0.5882, + "step": 2857 + }, + { + "epoch": 0.6793439505585929, + "grad_norm": 1.2033813292005264, + "learning_rate": 2.4634140236095284e-06, + "loss": 0.5317, + "step": 2858 + }, + { + "epoch": 0.6795816496315664, + "grad_norm": 1.3192266857835881, + "learning_rate": 2.4600970099733684e-06, + "loss": 0.5683, + "step": 2859 + }, + { + "epoch": 0.6798193487045401, + "grad_norm": 1.4101109866000088, + "learning_rate": 2.456781502238574e-06, + "loss": 0.5629, + "step": 2860 + }, + { + "epoch": 0.6800570477775136, + "grad_norm": 1.1543979011646313, + "learning_rate": 2.4534675023709004e-06, + "loss": 0.5551, + "step": 2861 + }, + { + "epoch": 0.6802947468504873, + "grad_norm": 1.1846254863385617, + "learning_rate": 2.4501550123352105e-06, + "loss": 0.5806, + "step": 2862 + }, + { + "epoch": 0.6805324459234608, + "grad_norm": 1.776974123769363, + "learning_rate": 2.4468440340954664e-06, + "loss": 0.6006, + "step": 2863 + }, + { + "epoch": 0.6807701449964345, + "grad_norm": 1.4547668409965302, + "learning_rate": 2.4435345696147404e-06, + "loss": 0.5845, + "step": 2864 + }, + { + "epoch": 0.6810078440694082, + "grad_norm": 1.2002930056447905, + "learning_rate": 2.4402266208552046e-06, + "loss": 0.5613, + "step": 2865 + }, + { + "epoch": 0.6812455431423817, + "grad_norm": 1.6074887439369634, + "learning_rate": 2.436920189778134e-06, + "loss": 0.5647, + "step": 2866 + }, + { + "epoch": 0.6814832422153554, + "grad_norm": 1.6822952576629444, + "learning_rate": 2.4336152783438984e-06, + "loss": 0.6061, + "step": 2867 + }, + { + "epoch": 0.6817209412883289, + "grad_norm": 1.2976528886043375, + "learning_rate": 2.4303118885119727e-06, + "loss": 0.5367, + "step": 2868 + }, + { + "epoch": 0.6819586403613026, + "grad_norm": 2.1035423684221395, + "learning_rate": 2.4270100222409275e-06, + "loss": 0.6062, + "step": 2869 + }, + { + "epoch": 0.6821963394342762, + "grad_norm": 1.4399468456290323, + "learning_rate": 2.423709681488431e-06, + "loss": 0.5487, + "step": 2870 + }, + { + "epoch": 0.6824340385072498, + "grad_norm": 1.19351394784005, + "learning_rate": 2.4204108682112443e-06, + "loss": 0.5426, + "step": 2871 + }, + { + "epoch": 0.6826717375802235, + "grad_norm": 1.2249309880381214, + "learning_rate": 2.4171135843652256e-06, + "loss": 0.5811, + "step": 2872 + }, + { + "epoch": 0.682909436653197, + "grad_norm": 1.5149776668003256, + "learning_rate": 2.4138178319053262e-06, + "loss": 0.5404, + "step": 2873 + }, + { + "epoch": 0.6831471357261707, + "grad_norm": 1.3429959947834988, + "learning_rate": 2.410523612785586e-06, + "loss": 0.5644, + "step": 2874 + }, + { + "epoch": 0.6833848347991442, + "grad_norm": 1.1434719947370102, + "learning_rate": 2.4072309289591394e-06, + "loss": 0.5513, + "step": 2875 + }, + { + "epoch": 0.6836225338721179, + "grad_norm": 1.3007484688562125, + "learning_rate": 2.4039397823782096e-06, + "loss": 0.5678, + "step": 2876 + }, + { + "epoch": 0.6838602329450915, + "grad_norm": 1.4679502068374575, + "learning_rate": 2.4006501749941097e-06, + "loss": 0.5545, + "step": 2877 + }, + { + "epoch": 0.6840979320180651, + "grad_norm": 1.2176596525807573, + "learning_rate": 2.397362108757236e-06, + "loss": 0.5074, + "step": 2878 + }, + { + "epoch": 0.6843356310910388, + "grad_norm": 1.2618747515034252, + "learning_rate": 2.3940755856170744e-06, + "loss": 0.5728, + "step": 2879 + }, + { + "epoch": 0.6845733301640123, + "grad_norm": 1.3927218688618288, + "learning_rate": 2.3907906075221953e-06, + "loss": 0.5333, + "step": 2880 + }, + { + "epoch": 0.684811029236986, + "grad_norm": 1.188418429998516, + "learning_rate": 2.387507176420256e-06, + "loss": 0.5574, + "step": 2881 + }, + { + "epoch": 0.6850487283099596, + "grad_norm": 1.509681238289089, + "learning_rate": 2.384225294257989e-06, + "loss": 0.5395, + "step": 2882 + }, + { + "epoch": 0.6852864273829332, + "grad_norm": 1.3145879038378638, + "learning_rate": 2.3809449629812154e-06, + "loss": 0.6208, + "step": 2883 + }, + { + "epoch": 0.6855241264559068, + "grad_norm": 1.1727638728264813, + "learning_rate": 2.3776661845348342e-06, + "loss": 0.5168, + "step": 2884 + }, + { + "epoch": 0.6857618255288804, + "grad_norm": 1.2719947867353576, + "learning_rate": 2.3743889608628253e-06, + "loss": 0.5371, + "step": 2885 + }, + { + "epoch": 0.6859995246018541, + "grad_norm": 1.1775861873661047, + "learning_rate": 2.371113293908242e-06, + "loss": 0.5395, + "step": 2886 + }, + { + "epoch": 0.6862372236748276, + "grad_norm": 1.2440517981606498, + "learning_rate": 2.3678391856132203e-06, + "loss": 0.5748, + "step": 2887 + }, + { + "epoch": 0.6864749227478013, + "grad_norm": 1.1790381314421488, + "learning_rate": 2.3645666379189704e-06, + "loss": 0.5764, + "step": 2888 + }, + { + "epoch": 0.6867126218207749, + "grad_norm": 1.3200148504532911, + "learning_rate": 2.3612956527657767e-06, + "loss": 0.5155, + "step": 2889 + }, + { + "epoch": 0.6869503208937485, + "grad_norm": 1.169340994505188, + "learning_rate": 2.358026232092995e-06, + "loss": 0.5707, + "step": 2890 + }, + { + "epoch": 0.6871880199667221, + "grad_norm": 1.2045187875579288, + "learning_rate": 2.3547583778390564e-06, + "loss": 0.5738, + "step": 2891 + }, + { + "epoch": 0.6874257190396957, + "grad_norm": 1.3520387265020841, + "learning_rate": 2.3514920919414636e-06, + "loss": 0.5211, + "step": 2892 + }, + { + "epoch": 0.6876634181126694, + "grad_norm": 2.0694736817412602, + "learning_rate": 2.348227376336789e-06, + "loss": 0.5835, + "step": 2893 + }, + { + "epoch": 0.687901117185643, + "grad_norm": 1.4351400760003536, + "learning_rate": 2.3449642329606707e-06, + "loss": 0.5906, + "step": 2894 + }, + { + "epoch": 0.6881388162586166, + "grad_norm": 1.2458420477573071, + "learning_rate": 2.341702663747819e-06, + "loss": 0.5892, + "step": 2895 + }, + { + "epoch": 0.6883765153315902, + "grad_norm": 1.6644437613478085, + "learning_rate": 2.338442670632009e-06, + "loss": 0.5706, + "step": 2896 + }, + { + "epoch": 0.6886142144045638, + "grad_norm": 1.5740604051703957, + "learning_rate": 2.335184255546083e-06, + "loss": 0.5614, + "step": 2897 + }, + { + "epoch": 0.6888519134775375, + "grad_norm": 1.2906751902926543, + "learning_rate": 2.3319274204219427e-06, + "loss": 0.5505, + "step": 2898 + }, + { + "epoch": 0.689089612550511, + "grad_norm": 1.381039300587738, + "learning_rate": 2.328672167190558e-06, + "loss": 0.5358, + "step": 2899 + }, + { + "epoch": 0.6893273116234847, + "grad_norm": 1.260138956741678, + "learning_rate": 2.3254184977819604e-06, + "loss": 0.5571, + "step": 2900 + }, + { + "epoch": 0.6893273116234847, + "eval_loss": 0.5681800246238708, + "eval_runtime": 278.6712, + "eval_samples_per_second": 39.035, + "eval_steps_per_second": 1.22, + "step": 2900 + }, + { + "epoch": 0.6895650106964583, + "grad_norm": 1.2743618518966204, + "learning_rate": 2.322166414125242e-06, + "loss": 0.6002, + "step": 2901 + }, + { + "epoch": 0.6898027097694319, + "grad_norm": 1.2134527172152756, + "learning_rate": 2.3189159181485517e-06, + "loss": 0.5679, + "step": 2902 + }, + { + "epoch": 0.6900404088424055, + "grad_norm": 1.2503267495636459, + "learning_rate": 2.3156670117790996e-06, + "loss": 0.5511, + "step": 2903 + }, + { + "epoch": 0.6902781079153791, + "grad_norm": 1.2509202355801665, + "learning_rate": 2.312419696943155e-06, + "loss": 0.5759, + "step": 2904 + }, + { + "epoch": 0.6905158069883528, + "grad_norm": 1.1454579932226538, + "learning_rate": 2.3091739755660425e-06, + "loss": 0.6132, + "step": 2905 + }, + { + "epoch": 0.6907535060613264, + "grad_norm": 1.262988373991108, + "learning_rate": 2.3059298495721377e-06, + "loss": 0.5809, + "step": 2906 + }, + { + "epoch": 0.6909912051343, + "grad_norm": 1.2281296364822565, + "learning_rate": 2.302687320884876e-06, + "loss": 0.5264, + "step": 2907 + }, + { + "epoch": 0.6912289042072736, + "grad_norm": 1.2303026691157126, + "learning_rate": 2.2994463914267435e-06, + "loss": 0.5418, + "step": 2908 + }, + { + "epoch": 0.6914666032802472, + "grad_norm": 1.6297581380195565, + "learning_rate": 2.2962070631192796e-06, + "loss": 0.542, + "step": 2909 + }, + { + "epoch": 0.6917043023532208, + "grad_norm": 1.183285288178771, + "learning_rate": 2.292969337883069e-06, + "loss": 0.571, + "step": 2910 + }, + { + "epoch": 0.6919420014261944, + "grad_norm": 1.1322822351609283, + "learning_rate": 2.289733217637753e-06, + "loss": 0.522, + "step": 2911 + }, + { + "epoch": 0.6921797004991681, + "grad_norm": 1.1593642608166748, + "learning_rate": 2.2864987043020176e-06, + "loss": 0.5848, + "step": 2912 + }, + { + "epoch": 0.6924173995721417, + "grad_norm": 1.2758173670690551, + "learning_rate": 2.2832657997935982e-06, + "loss": 0.5095, + "step": 2913 + }, + { + "epoch": 0.6926550986451153, + "grad_norm": 1.218121682148563, + "learning_rate": 2.2800345060292716e-06, + "loss": 0.5645, + "step": 2914 + }, + { + "epoch": 0.6928927977180889, + "grad_norm": 1.4819715019855433, + "learning_rate": 2.2768048249248648e-06, + "loss": 0.5589, + "step": 2915 + }, + { + "epoch": 0.6931304967910625, + "grad_norm": 1.22948699557779, + "learning_rate": 2.273576758395246e-06, + "loss": 0.5605, + "step": 2916 + }, + { + "epoch": 0.6933681958640361, + "grad_norm": 1.1805982358766494, + "learning_rate": 2.2703503083543288e-06, + "loss": 0.5937, + "step": 2917 + }, + { + "epoch": 0.6936058949370097, + "grad_norm": 1.136260242096098, + "learning_rate": 2.267125476715063e-06, + "loss": 0.538, + "step": 2918 + }, + { + "epoch": 0.6938435940099834, + "grad_norm": 1.22671239143086, + "learning_rate": 2.2639022653894443e-06, + "loss": 0.5153, + "step": 2919 + }, + { + "epoch": 0.694081293082957, + "grad_norm": 1.6077521744736338, + "learning_rate": 2.2606806762885054e-06, + "loss": 0.5688, + "step": 2920 + }, + { + "epoch": 0.6943189921559306, + "grad_norm": 1.2508811710955028, + "learning_rate": 2.2574607113223184e-06, + "loss": 0.5496, + "step": 2921 + }, + { + "epoch": 0.6945566912289042, + "grad_norm": 1.518406788408734, + "learning_rate": 2.2542423723999896e-06, + "loss": 0.5299, + "step": 2922 + }, + { + "epoch": 0.6947943903018778, + "grad_norm": 1.214955112181534, + "learning_rate": 2.2510256614296638e-06, + "loss": 0.5579, + "step": 2923 + }, + { + "epoch": 0.6950320893748514, + "grad_norm": 2.141838156162841, + "learning_rate": 2.24781058031852e-06, + "loss": 0.5119, + "step": 2924 + }, + { + "epoch": 0.695269788447825, + "grad_norm": 1.210425327352138, + "learning_rate": 2.2445971309727725e-06, + "loss": 0.5021, + "step": 2925 + }, + { + "epoch": 0.6955074875207987, + "grad_norm": 1.3906518276978634, + "learning_rate": 2.241385315297664e-06, + "loss": 0.5763, + "step": 2926 + }, + { + "epoch": 0.6957451865937723, + "grad_norm": 1.1639517026189996, + "learning_rate": 2.238175135197471e-06, + "loss": 0.6134, + "step": 2927 + }, + { + "epoch": 0.6959828856667459, + "grad_norm": 2.1279453719525643, + "learning_rate": 2.2349665925755014e-06, + "loss": 0.545, + "step": 2928 + }, + { + "epoch": 0.6962205847397195, + "grad_norm": 1.4657223300477322, + "learning_rate": 2.2317596893340924e-06, + "loss": 0.5189, + "step": 2929 + }, + { + "epoch": 0.6964582838126931, + "grad_norm": 1.3642036759878955, + "learning_rate": 2.228554427374605e-06, + "loss": 0.5595, + "step": 2930 + }, + { + "epoch": 0.6966959828856667, + "grad_norm": 1.218803694378054, + "learning_rate": 2.225350808597432e-06, + "loss": 0.5308, + "step": 2931 + }, + { + "epoch": 0.6969336819586404, + "grad_norm": 1.4334418805955895, + "learning_rate": 2.2221488349019903e-06, + "loss": 0.5489, + "step": 2932 + }, + { + "epoch": 0.697171381031614, + "grad_norm": 1.1706632856321182, + "learning_rate": 2.218948508186719e-06, + "loss": 0.475, + "step": 2933 + }, + { + "epoch": 0.6974090801045876, + "grad_norm": 1.2213509130684097, + "learning_rate": 2.215749830349084e-06, + "loss": 0.559, + "step": 2934 + }, + { + "epoch": 0.6976467791775612, + "grad_norm": 1.4049414470778998, + "learning_rate": 2.2125528032855727e-06, + "loss": 0.5694, + "step": 2935 + }, + { + "epoch": 0.6978844782505348, + "grad_norm": 1.188415322572375, + "learning_rate": 2.2093574288916946e-06, + "loss": 0.5298, + "step": 2936 + }, + { + "epoch": 0.6981221773235085, + "grad_norm": 1.0729422385472756, + "learning_rate": 2.206163709061976e-06, + "loss": 0.5325, + "step": 2937 + }, + { + "epoch": 0.698359876396482, + "grad_norm": 1.1873974446728723, + "learning_rate": 2.2029716456899645e-06, + "loss": 0.5564, + "step": 2938 + }, + { + "epoch": 0.6985975754694557, + "grad_norm": 1.5073407695340426, + "learning_rate": 2.1997812406682266e-06, + "loss": 0.5477, + "step": 2939 + }, + { + "epoch": 0.6988352745424293, + "grad_norm": 1.4635176192858388, + "learning_rate": 2.196592495888347e-06, + "loss": 0.5802, + "step": 2940 + }, + { + "epoch": 0.6990729736154029, + "grad_norm": 1.2682801405629767, + "learning_rate": 2.1934054132409183e-06, + "loss": 0.5494, + "step": 2941 + }, + { + "epoch": 0.6993106726883765, + "grad_norm": 1.552293282278238, + "learning_rate": 2.1902199946155555e-06, + "loss": 0.6119, + "step": 2942 + }, + { + "epoch": 0.6995483717613501, + "grad_norm": 1.1825435985560317, + "learning_rate": 2.1870362419008844e-06, + "loss": 0.5326, + "step": 2943 + }, + { + "epoch": 0.6997860708343238, + "grad_norm": 1.7285957557781395, + "learning_rate": 2.183854156984545e-06, + "loss": 0.5778, + "step": 2944 + }, + { + "epoch": 0.7000237699072973, + "grad_norm": 1.2143870668312673, + "learning_rate": 2.1806737417531826e-06, + "loss": 0.5446, + "step": 2945 + }, + { + "epoch": 0.700261468980271, + "grad_norm": 1.6253634307580322, + "learning_rate": 2.1774949980924585e-06, + "loss": 0.5451, + "step": 2946 + }, + { + "epoch": 0.7004991680532446, + "grad_norm": 1.2935231723526814, + "learning_rate": 2.174317927887041e-06, + "loss": 0.5836, + "step": 2947 + }, + { + "epoch": 0.7007368671262182, + "grad_norm": 1.2778563592595138, + "learning_rate": 2.171142533020607e-06, + "loss": 0.54, + "step": 2948 + }, + { + "epoch": 0.7009745661991919, + "grad_norm": 1.2717787145553763, + "learning_rate": 2.1679688153758373e-06, + "loss": 0.5172, + "step": 2949 + }, + { + "epoch": 0.7012122652721654, + "grad_norm": 1.414439308128543, + "learning_rate": 2.1647967768344204e-06, + "loss": 0.5521, + "step": 2950 + }, + { + "epoch": 0.7014499643451391, + "grad_norm": 1.9501066891139769, + "learning_rate": 2.1616264192770496e-06, + "loss": 0.5665, + "step": 2951 + }, + { + "epoch": 0.7016876634181126, + "grad_norm": 1.4068129548076287, + "learning_rate": 2.1584577445834234e-06, + "loss": 0.5411, + "step": 2952 + }, + { + "epoch": 0.7019253624910863, + "grad_norm": 1.316219847953352, + "learning_rate": 2.1552907546322356e-06, + "loss": 0.55, + "step": 2953 + }, + { + "epoch": 0.7021630615640599, + "grad_norm": 1.1187254481243074, + "learning_rate": 2.152125451301188e-06, + "loss": 0.5885, + "step": 2954 + }, + { + "epoch": 0.7024007606370335, + "grad_norm": 1.2093587473355791, + "learning_rate": 2.14896183646698e-06, + "loss": 0.5393, + "step": 2955 + }, + { + "epoch": 0.7026384597100072, + "grad_norm": 1.1788166856862319, + "learning_rate": 2.1457999120053125e-06, + "loss": 0.5596, + "step": 2956 + }, + { + "epoch": 0.7028761587829807, + "grad_norm": 1.4917797068696286, + "learning_rate": 2.1426396797908764e-06, + "loss": 0.5397, + "step": 2957 + }, + { + "epoch": 0.7031138578559544, + "grad_norm": 1.2998051774616508, + "learning_rate": 2.139481141697368e-06, + "loss": 0.5254, + "step": 2958 + }, + { + "epoch": 0.7033515569289279, + "grad_norm": 1.3021979760945492, + "learning_rate": 2.136324299597474e-06, + "loss": 0.5931, + "step": 2959 + }, + { + "epoch": 0.7035892560019016, + "grad_norm": 1.27144721339873, + "learning_rate": 2.133169155362881e-06, + "loss": 0.5081, + "step": 2960 + }, + { + "epoch": 0.7038269550748752, + "grad_norm": 1.3015664423193956, + "learning_rate": 2.13001571086426e-06, + "loss": 0.5859, + "step": 2961 + }, + { + "epoch": 0.7040646541478488, + "grad_norm": 1.3471683535664603, + "learning_rate": 2.1268639679712814e-06, + "loss": 0.5711, + "step": 2962 + }, + { + "epoch": 0.7043023532208225, + "grad_norm": 1.2098472874032669, + "learning_rate": 2.1237139285526058e-06, + "loss": 0.5372, + "step": 2963 + }, + { + "epoch": 0.704540052293796, + "grad_norm": 1.4790409934849147, + "learning_rate": 2.1205655944758834e-06, + "loss": 0.5481, + "step": 2964 + }, + { + "epoch": 0.7047777513667697, + "grad_norm": 1.3160735215808004, + "learning_rate": 2.11741896760775e-06, + "loss": 0.6323, + "step": 2965 + }, + { + "epoch": 0.7050154504397432, + "grad_norm": 1.281571951864179, + "learning_rate": 2.1142740498138327e-06, + "loss": 0.6044, + "step": 2966 + }, + { + "epoch": 0.7052531495127169, + "grad_norm": 1.192321013300713, + "learning_rate": 2.1111308429587446e-06, + "loss": 0.4922, + "step": 2967 + }, + { + "epoch": 0.7054908485856906, + "grad_norm": 1.3020514048348912, + "learning_rate": 2.107989348906087e-06, + "loss": 0.5454, + "step": 2968 + }, + { + "epoch": 0.7057285476586641, + "grad_norm": 1.3246311645463134, + "learning_rate": 2.1048495695184382e-06, + "loss": 0.5828, + "step": 2969 + }, + { + "epoch": 0.7059662467316378, + "grad_norm": 1.1895428935456933, + "learning_rate": 2.1017115066573668e-06, + "loss": 0.5421, + "step": 2970 + }, + { + "epoch": 0.7062039458046113, + "grad_norm": 1.1853153248301067, + "learning_rate": 2.098575162183422e-06, + "loss": 0.5414, + "step": 2971 + }, + { + "epoch": 0.706441644877585, + "grad_norm": 1.2228898112284892, + "learning_rate": 2.0954405379561345e-06, + "loss": 0.5793, + "step": 2972 + }, + { + "epoch": 0.7066793439505586, + "grad_norm": 1.8684906198648799, + "learning_rate": 2.0923076358340112e-06, + "loss": 0.529, + "step": 2973 + }, + { + "epoch": 0.7069170430235322, + "grad_norm": 1.1866586560869163, + "learning_rate": 2.0891764576745426e-06, + "loss": 0.5197, + "step": 2974 + }, + { + "epoch": 0.7071547420965059, + "grad_norm": 1.3456864811704945, + "learning_rate": 2.0860470053341957e-06, + "loss": 0.5339, + "step": 2975 + }, + { + "epoch": 0.7073924411694794, + "grad_norm": 1.400100336497434, + "learning_rate": 2.0829192806684158e-06, + "loss": 0.5454, + "step": 2976 + }, + { + "epoch": 0.7076301402424531, + "grad_norm": 1.287212071010978, + "learning_rate": 2.0797932855316183e-06, + "loss": 0.5456, + "step": 2977 + }, + { + "epoch": 0.7078678393154266, + "grad_norm": 1.4545528094534008, + "learning_rate": 2.0766690217771986e-06, + "loss": 0.5538, + "step": 2978 + }, + { + "epoch": 0.7081055383884003, + "grad_norm": 1.2730815560962423, + "learning_rate": 2.073546491257524e-06, + "loss": 0.5367, + "step": 2979 + }, + { + "epoch": 0.708343237461374, + "grad_norm": 1.127002865793415, + "learning_rate": 2.070425695823936e-06, + "loss": 0.6009, + "step": 2980 + }, + { + "epoch": 0.7085809365343475, + "grad_norm": 1.5347018248698554, + "learning_rate": 2.0673066373267425e-06, + "loss": 0.5824, + "step": 2981 + }, + { + "epoch": 0.7088186356073212, + "grad_norm": 1.3073717386823567, + "learning_rate": 2.064189317615225e-06, + "loss": 0.5392, + "step": 2982 + }, + { + "epoch": 0.7090563346802947, + "grad_norm": 1.1594117025382047, + "learning_rate": 2.061073738537635e-06, + "loss": 0.5294, + "step": 2983 + }, + { + "epoch": 0.7092940337532684, + "grad_norm": 1.171728657662454, + "learning_rate": 2.057959901941191e-06, + "loss": 0.6184, + "step": 2984 + }, + { + "epoch": 0.7095317328262419, + "grad_norm": 1.3550605534124591, + "learning_rate": 2.054847809672076e-06, + "loss": 0.4872, + "step": 2985 + }, + { + "epoch": 0.7097694318992156, + "grad_norm": 1.327681716755177, + "learning_rate": 2.051737463575441e-06, + "loss": 0.564, + "step": 2986 + }, + { + "epoch": 0.7100071309721893, + "grad_norm": 1.629375084096743, + "learning_rate": 2.048628865495403e-06, + "loss": 0.5203, + "step": 2987 + }, + { + "epoch": 0.7102448300451628, + "grad_norm": 1.2321022871109746, + "learning_rate": 2.0455220172750416e-06, + "loss": 0.5237, + "step": 2988 + }, + { + "epoch": 0.7104825291181365, + "grad_norm": 1.1477777043751685, + "learning_rate": 2.0424169207563954e-06, + "loss": 0.5903, + "step": 2989 + }, + { + "epoch": 0.71072022819111, + "grad_norm": 1.255355278105762, + "learning_rate": 2.039313577780469e-06, + "loss": 0.62, + "step": 2990 + }, + { + "epoch": 0.7109579272640837, + "grad_norm": 1.4975773193567874, + "learning_rate": 2.0362119901872262e-06, + "loss": 0.5965, + "step": 2991 + }, + { + "epoch": 0.7111956263370572, + "grad_norm": 1.1705090392548188, + "learning_rate": 2.0331121598155905e-06, + "loss": 0.5611, + "step": 2992 + }, + { + "epoch": 0.7114333254100309, + "grad_norm": 1.1424623086463552, + "learning_rate": 2.0300140885034403e-06, + "loss": 0.5774, + "step": 2993 + }, + { + "epoch": 0.7116710244830046, + "grad_norm": 1.1991748278339214, + "learning_rate": 2.026917778087614e-06, + "loss": 0.4975, + "step": 2994 + }, + { + "epoch": 0.7119087235559781, + "grad_norm": 1.2444550434903148, + "learning_rate": 2.023823230403907e-06, + "loss": 0.5846, + "step": 2995 + }, + { + "epoch": 0.7121464226289518, + "grad_norm": 1.1843349341853713, + "learning_rate": 2.0207304472870654e-06, + "loss": 0.5518, + "step": 2996 + }, + { + "epoch": 0.7123841217019253, + "grad_norm": 1.3024952749719423, + "learning_rate": 2.017639430570794e-06, + "loss": 0.5941, + "step": 2997 + }, + { + "epoch": 0.712621820774899, + "grad_norm": 2.863422292337457, + "learning_rate": 2.0145501820877468e-06, + "loss": 0.5481, + "step": 2998 + }, + { + "epoch": 0.7128595198478725, + "grad_norm": 1.1410483719753455, + "learning_rate": 2.011462703669532e-06, + "loss": 0.5237, + "step": 2999 + }, + { + "epoch": 0.7130972189208462, + "grad_norm": 1.2011094014772845, + "learning_rate": 2.008376997146705e-06, + "loss": 0.5254, + "step": 3000 + }, + { + "epoch": 0.7130972189208462, + "eval_loss": 0.5651676654815674, + "eval_runtime": 278.6133, + "eval_samples_per_second": 39.043, + "eval_steps_per_second": 1.22, + "step": 3000 + }, + { + "epoch": 0.7133349179938199, + "grad_norm": 1.1391684454323439, + "learning_rate": 2.005293064348773e-06, + "loss": 0.5587, + "step": 3001 + }, + { + "epoch": 0.7135726170667934, + "grad_norm": 1.175039374719243, + "learning_rate": 2.0022109071041905e-06, + "loss": 0.5293, + "step": 3002 + }, + { + "epoch": 0.7138103161397671, + "grad_norm": 1.331858891368864, + "learning_rate": 1.9991305272403637e-06, + "loss": 0.5656, + "step": 3003 + }, + { + "epoch": 0.7140480152127406, + "grad_norm": 1.1723680079449648, + "learning_rate": 1.996051926583636e-06, + "loss": 0.5612, + "step": 3004 + }, + { + "epoch": 0.7142857142857143, + "grad_norm": 1.2651238796064777, + "learning_rate": 1.9929751069593033e-06, + "loss": 0.5724, + "step": 3005 + }, + { + "epoch": 0.7145234133586879, + "grad_norm": 1.850700099271677, + "learning_rate": 1.9899000701916033e-06, + "loss": 0.5437, + "step": 3006 + }, + { + "epoch": 0.7147611124316615, + "grad_norm": 1.2310418566231582, + "learning_rate": 1.9868268181037186e-06, + "loss": 0.5985, + "step": 3007 + }, + { + "epoch": 0.7149988115046352, + "grad_norm": 1.2251273231863045, + "learning_rate": 1.983755352517768e-06, + "loss": 0.5829, + "step": 3008 + }, + { + "epoch": 0.7152365105776087, + "grad_norm": 1.2443679038770428, + "learning_rate": 1.9806856752548178e-06, + "loss": 0.5428, + "step": 3009 + }, + { + "epoch": 0.7154742096505824, + "grad_norm": 1.2522340351296133, + "learning_rate": 1.977617788134869e-06, + "loss": 0.5303, + "step": 3010 + }, + { + "epoch": 0.715711908723556, + "grad_norm": 1.8198172968095345, + "learning_rate": 1.974551692976866e-06, + "loss": 0.4914, + "step": 3011 + }, + { + "epoch": 0.7159496077965296, + "grad_norm": 1.119211258932051, + "learning_rate": 1.9714873915986848e-06, + "loss": 0.5535, + "step": 3012 + }, + { + "epoch": 0.7161873068695032, + "grad_norm": 1.250620132976224, + "learning_rate": 1.968424885817143e-06, + "loss": 0.5813, + "step": 3013 + }, + { + "epoch": 0.7164250059424768, + "grad_norm": 1.1520938032042989, + "learning_rate": 1.965364177447991e-06, + "loss": 0.5073, + "step": 3014 + }, + { + "epoch": 0.7166627050154505, + "grad_norm": 1.2103794123510823, + "learning_rate": 1.9623052683059164e-06, + "loss": 0.5844, + "step": 3015 + }, + { + "epoch": 0.716900404088424, + "grad_norm": 1.183235062803906, + "learning_rate": 1.959248160204534e-06, + "loss": 0.5084, + "step": 3016 + }, + { + "epoch": 0.7171381031613977, + "grad_norm": 1.1732113611620743, + "learning_rate": 1.956192854956397e-06, + "loss": 0.5488, + "step": 3017 + }, + { + "epoch": 0.7173758022343713, + "grad_norm": 1.1477439661632367, + "learning_rate": 1.9531393543729865e-06, + "loss": 0.5907, + "step": 3018 + }, + { + "epoch": 0.7176135013073449, + "grad_norm": 1.2093451497008603, + "learning_rate": 1.9500876602647167e-06, + "loss": 0.5694, + "step": 3019 + }, + { + "epoch": 0.7178512003803185, + "grad_norm": 1.2518076564051352, + "learning_rate": 1.9470377744409253e-06, + "loss": 0.5471, + "step": 3020 + }, + { + "epoch": 0.7180888994532921, + "grad_norm": 1.2852812183839475, + "learning_rate": 1.943989698709882e-06, + "loss": 0.5494, + "step": 3021 + }, + { + "epoch": 0.7183265985262658, + "grad_norm": 1.1791854725589512, + "learning_rate": 1.9409434348787824e-06, + "loss": 0.4843, + "step": 3022 + }, + { + "epoch": 0.7185642975992393, + "grad_norm": 1.2287169533472668, + "learning_rate": 1.937898984753751e-06, + "loss": 0.5588, + "step": 3023 + }, + { + "epoch": 0.718801996672213, + "grad_norm": 1.425882276251666, + "learning_rate": 1.9348563501398288e-06, + "loss": 0.5715, + "step": 3024 + }, + { + "epoch": 0.7190396957451866, + "grad_norm": 2.1015074810065277, + "learning_rate": 1.931815532840987e-06, + "loss": 0.5305, + "step": 3025 + }, + { + "epoch": 0.7192773948181602, + "grad_norm": 1.1899085918296417, + "learning_rate": 1.9287765346601185e-06, + "loss": 0.54, + "step": 3026 + }, + { + "epoch": 0.7195150938911338, + "grad_norm": 1.3673567486571454, + "learning_rate": 1.925739357399038e-06, + "loss": 0.5536, + "step": 3027 + }, + { + "epoch": 0.7197527929641074, + "grad_norm": 1.3635272384342083, + "learning_rate": 1.922704002858476e-06, + "loss": 0.5663, + "step": 3028 + }, + { + "epoch": 0.7199904920370811, + "grad_norm": 1.3565705922802063, + "learning_rate": 1.919670472838088e-06, + "loss": 0.6042, + "step": 3029 + }, + { + "epoch": 0.7202281911100546, + "grad_norm": 1.3212448810425161, + "learning_rate": 1.9166387691364442e-06, + "loss": 0.5478, + "step": 3030 + }, + { + "epoch": 0.7204658901830283, + "grad_norm": 1.3308641651324886, + "learning_rate": 1.913608893551036e-06, + "loss": 0.5879, + "step": 3031 + }, + { + "epoch": 0.7207035892560019, + "grad_norm": 1.3240496024391093, + "learning_rate": 1.9105808478782644e-06, + "loss": 0.5717, + "step": 3032 + }, + { + "epoch": 0.7209412883289755, + "grad_norm": 1.2481219441101794, + "learning_rate": 1.907554633913451e-06, + "loss": 0.5239, + "step": 3033 + }, + { + "epoch": 0.7211789874019491, + "grad_norm": 1.654943793287123, + "learning_rate": 1.9045302534508298e-06, + "loss": 0.5893, + "step": 3034 + }, + { + "epoch": 0.7214166864749227, + "grad_norm": 1.232381314673576, + "learning_rate": 1.9015077082835487e-06, + "loss": 0.5726, + "step": 3035 + }, + { + "epoch": 0.7216543855478964, + "grad_norm": 1.3396926583408935, + "learning_rate": 1.8984870002036627e-06, + "loss": 0.513, + "step": 3036 + }, + { + "epoch": 0.72189208462087, + "grad_norm": 1.3025131062701227, + "learning_rate": 1.8954681310021434e-06, + "loss": 0.5258, + "step": 3037 + }, + { + "epoch": 0.7221297836938436, + "grad_norm": 1.321021481305135, + "learning_rate": 1.8924511024688702e-06, + "loss": 0.5595, + "step": 3038 + }, + { + "epoch": 0.7223674827668172, + "grad_norm": 1.1921919429633083, + "learning_rate": 1.8894359163926312e-06, + "loss": 0.5557, + "step": 3039 + }, + { + "epoch": 0.7226051818397908, + "grad_norm": 1.476260334731851, + "learning_rate": 1.8864225745611197e-06, + "loss": 0.5918, + "step": 3040 + }, + { + "epoch": 0.7228428809127644, + "grad_norm": 1.1094846100784337, + "learning_rate": 1.8834110787609395e-06, + "loss": 0.5389, + "step": 3041 + }, + { + "epoch": 0.723080579985738, + "grad_norm": 1.3081150989111687, + "learning_rate": 1.8804014307775965e-06, + "loss": 0.57, + "step": 3042 + }, + { + "epoch": 0.7233182790587117, + "grad_norm": 1.2939218061574038, + "learning_rate": 1.8773936323955055e-06, + "loss": 0.5399, + "step": 3043 + }, + { + "epoch": 0.7235559781316853, + "grad_norm": 1.237271905094372, + "learning_rate": 1.8743876853979781e-06, + "loss": 0.5674, + "step": 3044 + }, + { + "epoch": 0.7237936772046589, + "grad_norm": 1.2200562740934084, + "learning_rate": 1.8713835915672328e-06, + "loss": 0.5118, + "step": 3045 + }, + { + "epoch": 0.7240313762776325, + "grad_norm": 1.1207240150947466, + "learning_rate": 1.8683813526843898e-06, + "loss": 0.4985, + "step": 3046 + }, + { + "epoch": 0.7242690753506061, + "grad_norm": 1.3748117257621784, + "learning_rate": 1.865380970529469e-06, + "loss": 0.5527, + "step": 3047 + }, + { + "epoch": 0.7245067744235798, + "grad_norm": 1.2540749232197617, + "learning_rate": 1.8623824468813844e-06, + "loss": 0.5873, + "step": 3048 + }, + { + "epoch": 0.7247444734965534, + "grad_norm": 1.3386907348375234, + "learning_rate": 1.8593857835179557e-06, + "loss": 0.5659, + "step": 3049 + }, + { + "epoch": 0.724982172569527, + "grad_norm": 1.9774649555403498, + "learning_rate": 1.856390982215895e-06, + "loss": 0.549, + "step": 3050 + }, + { + "epoch": 0.7252198716425006, + "grad_norm": 1.200898730317029, + "learning_rate": 1.8533980447508138e-06, + "loss": 0.5547, + "step": 3051 + }, + { + "epoch": 0.7254575707154742, + "grad_norm": 1.2322436351388026, + "learning_rate": 1.8504069728972124e-06, + "loss": 0.5992, + "step": 3052 + }, + { + "epoch": 0.7256952697884478, + "grad_norm": 1.1756297232718562, + "learning_rate": 1.8474177684284917e-06, + "loss": 0.5265, + "step": 3053 + }, + { + "epoch": 0.7259329688614214, + "grad_norm": 1.2886497294144292, + "learning_rate": 1.844430433116941e-06, + "loss": 0.5575, + "step": 3054 + }, + { + "epoch": 0.7261706679343951, + "grad_norm": 1.2155776745263205, + "learning_rate": 1.8414449687337467e-06, + "loss": 0.5319, + "step": 3055 + }, + { + "epoch": 0.7264083670073687, + "grad_norm": 1.3203544503757492, + "learning_rate": 1.8384613770489767e-06, + "loss": 0.5273, + "step": 3056 + }, + { + "epoch": 0.7266460660803423, + "grad_norm": 1.438330673592324, + "learning_rate": 1.8354796598315977e-06, + "loss": 0.5823, + "step": 3057 + }, + { + "epoch": 0.7268837651533159, + "grad_norm": 1.5367353450596926, + "learning_rate": 1.8324998188494608e-06, + "loss": 0.5332, + "step": 3058 + }, + { + "epoch": 0.7271214642262895, + "grad_norm": 1.2964042619175427, + "learning_rate": 1.8295218558693067e-06, + "loss": 0.5455, + "step": 3059 + }, + { + "epoch": 0.7273591632992631, + "grad_norm": 1.9118079496362554, + "learning_rate": 1.8265457726567582e-06, + "loss": 0.5331, + "step": 3060 + }, + { + "epoch": 0.7275968623722368, + "grad_norm": 1.2622115613391622, + "learning_rate": 1.8235715709763285e-06, + "loss": 0.5481, + "step": 3061 + }, + { + "epoch": 0.7278345614452104, + "grad_norm": 1.232256201347321, + "learning_rate": 1.8205992525914135e-06, + "loss": 0.5773, + "step": 3062 + }, + { + "epoch": 0.728072260518184, + "grad_norm": 1.2215314921767861, + "learning_rate": 1.8176288192642944e-06, + "loss": 0.6065, + "step": 3063 + }, + { + "epoch": 0.7283099595911576, + "grad_norm": 1.279743914672167, + "learning_rate": 1.8146602727561286e-06, + "loss": 0.5171, + "step": 3064 + }, + { + "epoch": 0.7285476586641312, + "grad_norm": 1.2807894167331684, + "learning_rate": 1.8116936148269609e-06, + "loss": 0.5385, + "step": 3065 + }, + { + "epoch": 0.7287853577371048, + "grad_norm": 1.4505752998266273, + "learning_rate": 1.8087288472357152e-06, + "loss": 0.5705, + "step": 3066 + }, + { + "epoch": 0.7290230568100784, + "grad_norm": 1.1524672544212755, + "learning_rate": 1.8057659717401948e-06, + "loss": 0.5571, + "step": 3067 + }, + { + "epoch": 0.7292607558830521, + "grad_norm": 1.2728908645800672, + "learning_rate": 1.8028049900970768e-06, + "loss": 0.5247, + "step": 3068 + }, + { + "epoch": 0.7294984549560257, + "grad_norm": 1.1264314083632943, + "learning_rate": 1.7998459040619215e-06, + "loss": 0.5465, + "step": 3069 + }, + { + "epoch": 0.7297361540289993, + "grad_norm": 1.5249015479071113, + "learning_rate": 1.7968887153891622e-06, + "loss": 0.5475, + "step": 3070 + }, + { + "epoch": 0.7299738531019729, + "grad_norm": 1.220375898617576, + "learning_rate": 1.7939334258321094e-06, + "loss": 0.5709, + "step": 3071 + }, + { + "epoch": 0.7302115521749465, + "grad_norm": 1.2003317952522732, + "learning_rate": 1.7909800371429425e-06, + "loss": 0.5372, + "step": 3072 + }, + { + "epoch": 0.7304492512479202, + "grad_norm": 1.2998936545639845, + "learning_rate": 1.7880285510727197e-06, + "loss": 0.5119, + "step": 3073 + }, + { + "epoch": 0.7306869503208937, + "grad_norm": 1.2620877391983294, + "learning_rate": 1.7850789693713682e-06, + "loss": 0.5511, + "step": 3074 + }, + { + "epoch": 0.7309246493938674, + "grad_norm": 1.1763957886598466, + "learning_rate": 1.7821312937876883e-06, + "loss": 0.5953, + "step": 3075 + }, + { + "epoch": 0.731162348466841, + "grad_norm": 1.255776153711314, + "learning_rate": 1.7791855260693458e-06, + "loss": 0.5117, + "step": 3076 + }, + { + "epoch": 0.7314000475398146, + "grad_norm": 1.3724146035977396, + "learning_rate": 1.7762416679628792e-06, + "loss": 0.5316, + "step": 3077 + }, + { + "epoch": 0.7316377466127882, + "grad_norm": 1.3193772804092707, + "learning_rate": 1.7732997212136943e-06, + "loss": 0.5539, + "step": 3078 + }, + { + "epoch": 0.7318754456857618, + "grad_norm": 1.3121093916043247, + "learning_rate": 1.7703596875660645e-06, + "loss": 0.5438, + "step": 3079 + }, + { + "epoch": 0.7321131447587355, + "grad_norm": 1.1854112560124905, + "learning_rate": 1.7674215687631247e-06, + "loss": 0.5811, + "step": 3080 + }, + { + "epoch": 0.732350843831709, + "grad_norm": 1.179812552813857, + "learning_rate": 1.7644853665468776e-06, + "loss": 0.5758, + "step": 3081 + }, + { + "epoch": 0.7325885429046827, + "grad_norm": 1.3455780094437981, + "learning_rate": 1.7615510826581906e-06, + "loss": 0.5373, + "step": 3082 + }, + { + "epoch": 0.7328262419776563, + "grad_norm": 1.3076188086363456, + "learning_rate": 1.758618718836793e-06, + "loss": 0.5568, + "step": 3083 + }, + { + "epoch": 0.7330639410506299, + "grad_norm": 1.3256896887915994, + "learning_rate": 1.755688276821273e-06, + "loss": 0.5555, + "step": 3084 + }, + { + "epoch": 0.7333016401236035, + "grad_norm": 1.1602081752370497, + "learning_rate": 1.7527597583490825e-06, + "loss": 0.5626, + "step": 3085 + }, + { + "epoch": 0.7335393391965771, + "grad_norm": 1.593525158745214, + "learning_rate": 1.7498331651565314e-06, + "loss": 0.5203, + "step": 3086 + }, + { + "epoch": 0.7337770382695508, + "grad_norm": 1.2970802371704404, + "learning_rate": 1.746908498978791e-06, + "loss": 0.5225, + "step": 3087 + }, + { + "epoch": 0.7340147373425243, + "grad_norm": 1.4038108611809534, + "learning_rate": 1.743985761549884e-06, + "loss": 0.5779, + "step": 3088 + }, + { + "epoch": 0.734252436415498, + "grad_norm": 1.1942505102871441, + "learning_rate": 1.7410649546026947e-06, + "loss": 0.5622, + "step": 3089 + }, + { + "epoch": 0.7344901354884716, + "grad_norm": 1.1372220779688857, + "learning_rate": 1.738146079868962e-06, + "loss": 0.5897, + "step": 3090 + }, + { + "epoch": 0.7347278345614452, + "grad_norm": 1.2998822836529749, + "learning_rate": 1.7352291390792798e-06, + "loss": 0.5685, + "step": 3091 + }, + { + "epoch": 0.7349655336344189, + "grad_norm": 1.182924693607127, + "learning_rate": 1.7323141339630905e-06, + "loss": 0.4789, + "step": 3092 + }, + { + "epoch": 0.7352032327073924, + "grad_norm": 1.1471998955496148, + "learning_rate": 1.7294010662486948e-06, + "loss": 0.5158, + "step": 3093 + }, + { + "epoch": 0.7354409317803661, + "grad_norm": 1.1192392610799589, + "learning_rate": 1.7264899376632415e-06, + "loss": 0.5078, + "step": 3094 + }, + { + "epoch": 0.7356786308533396, + "grad_norm": 1.2414417866986565, + "learning_rate": 1.7235807499327335e-06, + "loss": 0.5725, + "step": 3095 + }, + { + "epoch": 0.7359163299263133, + "grad_norm": 1.3547390798518557, + "learning_rate": 1.720673504782016e-06, + "loss": 0.5568, + "step": 3096 + }, + { + "epoch": 0.736154028999287, + "grad_norm": 1.134055547483995, + "learning_rate": 1.7177682039347875e-06, + "loss": 0.5545, + "step": 3097 + }, + { + "epoch": 0.7363917280722605, + "grad_norm": 1.23136097517721, + "learning_rate": 1.7148648491135945e-06, + "loss": 0.5227, + "step": 3098 + }, + { + "epoch": 0.7366294271452342, + "grad_norm": 1.1985465195813212, + "learning_rate": 1.7119634420398273e-06, + "loss": 0.5377, + "step": 3099 + }, + { + "epoch": 0.7368671262182077, + "grad_norm": 1.2074786882704827, + "learning_rate": 1.709063984433721e-06, + "loss": 0.5653, + "step": 3100 + }, + { + "epoch": 0.7368671262182077, + "eval_loss": 0.5626891255378723, + "eval_runtime": 278.8359, + "eval_samples_per_second": 39.012, + "eval_steps_per_second": 1.219, + "step": 3100 + }, + { + "epoch": 0.7371048252911814, + "grad_norm": 1.2554724321972963, + "learning_rate": 1.7061664780143555e-06, + "loss": 0.5179, + "step": 3101 + }, + { + "epoch": 0.7373425243641549, + "grad_norm": 1.4124639743860714, + "learning_rate": 1.7032709244996559e-06, + "loss": 0.5689, + "step": 3102 + }, + { + "epoch": 0.7375802234371286, + "grad_norm": 1.4201929718411148, + "learning_rate": 1.7003773256063882e-06, + "loss": 0.5715, + "step": 3103 + }, + { + "epoch": 0.7378179225101023, + "grad_norm": 1.369805731802236, + "learning_rate": 1.6974856830501567e-06, + "loss": 0.5788, + "step": 3104 + }, + { + "epoch": 0.7380556215830758, + "grad_norm": 1.1295989765130179, + "learning_rate": 1.6945959985454097e-06, + "loss": 0.526, + "step": 3105 + }, + { + "epoch": 0.7382933206560495, + "grad_norm": 1.400045974355714, + "learning_rate": 1.6917082738054319e-06, + "loss": 0.529, + "step": 3106 + }, + { + "epoch": 0.738531019729023, + "grad_norm": 1.362510875391624, + "learning_rate": 1.6888225105423505e-06, + "loss": 0.5306, + "step": 3107 + }, + { + "epoch": 0.7387687188019967, + "grad_norm": 1.1847472427714039, + "learning_rate": 1.685938710467122e-06, + "loss": 0.5349, + "step": 3108 + }, + { + "epoch": 0.7390064178749702, + "grad_norm": 1.2647871069226764, + "learning_rate": 1.6830568752895455e-06, + "loss": 0.5776, + "step": 3109 + }, + { + "epoch": 0.7392441169479439, + "grad_norm": 1.0600395685706605, + "learning_rate": 1.6801770067182526e-06, + "loss": 0.4888, + "step": 3110 + }, + { + "epoch": 0.7394818160209176, + "grad_norm": 2.2252531787639374, + "learning_rate": 1.6772991064607113e-06, + "loss": 0.554, + "step": 3111 + }, + { + "epoch": 0.7397195150938911, + "grad_norm": 1.360291769849451, + "learning_rate": 1.6744231762232178e-06, + "loss": 0.5774, + "step": 3112 + }, + { + "epoch": 0.7399572141668648, + "grad_norm": 1.0630517216600648, + "learning_rate": 1.6715492177109039e-06, + "loss": 0.4999, + "step": 3113 + }, + { + "epoch": 0.7401949132398383, + "grad_norm": 1.341639989720774, + "learning_rate": 1.6686772326277317e-06, + "loss": 0.6455, + "step": 3114 + }, + { + "epoch": 0.740432612312812, + "grad_norm": 1.1014157205047532, + "learning_rate": 1.6658072226764949e-06, + "loss": 0.5471, + "step": 3115 + }, + { + "epoch": 0.7406703113857855, + "grad_norm": 1.295638780772937, + "learning_rate": 1.6629391895588109e-06, + "loss": 0.5528, + "step": 3116 + }, + { + "epoch": 0.7409080104587592, + "grad_norm": 1.1696414528031163, + "learning_rate": 1.6600731349751303e-06, + "loss": 0.5789, + "step": 3117 + }, + { + "epoch": 0.7411457095317329, + "grad_norm": 1.2614058279443974, + "learning_rate": 1.6572090606247294e-06, + "loss": 0.5619, + "step": 3118 + }, + { + "epoch": 0.7413834086047064, + "grad_norm": 1.1680132135253132, + "learning_rate": 1.6543469682057105e-06, + "loss": 0.5403, + "step": 3119 + }, + { + "epoch": 0.7416211076776801, + "grad_norm": 1.2074998244713104, + "learning_rate": 1.6514868594149975e-06, + "loss": 0.5548, + "step": 3120 + }, + { + "epoch": 0.7418588067506536, + "grad_norm": 3.1555283122062483, + "learning_rate": 1.6486287359483422e-06, + "loss": 0.6291, + "step": 3121 + }, + { + "epoch": 0.7420965058236273, + "grad_norm": 1.2286164932575105, + "learning_rate": 1.6457725995003182e-06, + "loss": 0.6329, + "step": 3122 + }, + { + "epoch": 0.742334204896601, + "grad_norm": 1.452789334960498, + "learning_rate": 1.6429184517643222e-06, + "loss": 0.543, + "step": 3123 + }, + { + "epoch": 0.7425719039695745, + "grad_norm": 1.522775021949409, + "learning_rate": 1.6400662944325656e-06, + "loss": 0.5582, + "step": 3124 + }, + { + "epoch": 0.7428096030425482, + "grad_norm": 1.090185800896134, + "learning_rate": 1.6372161291960897e-06, + "loss": 0.5468, + "step": 3125 + }, + { + "epoch": 0.7430473021155217, + "grad_norm": 1.1413546085158677, + "learning_rate": 1.6343679577447492e-06, + "loss": 0.5838, + "step": 3126 + }, + { + "epoch": 0.7432850011884954, + "grad_norm": 1.2439144159555635, + "learning_rate": 1.6315217817672142e-06, + "loss": 0.5495, + "step": 3127 + }, + { + "epoch": 0.7435227002614689, + "grad_norm": 1.4552564051142858, + "learning_rate": 1.628677602950975e-06, + "loss": 0.5624, + "step": 3128 + }, + { + "epoch": 0.7437603993344426, + "grad_norm": 1.2960391545044614, + "learning_rate": 1.6258354229823381e-06, + "loss": 0.5469, + "step": 3129 + }, + { + "epoch": 0.7439980984074163, + "grad_norm": 1.3149974753712506, + "learning_rate": 1.6229952435464252e-06, + "loss": 0.539, + "step": 3130 + }, + { + "epoch": 0.7442357974803898, + "grad_norm": 1.3027311395724246, + "learning_rate": 1.6201570663271682e-06, + "loss": 0.5736, + "step": 3131 + }, + { + "epoch": 0.7444734965533635, + "grad_norm": 1.90291901534616, + "learning_rate": 1.6173208930073152e-06, + "loss": 0.562, + "step": 3132 + }, + { + "epoch": 0.744711195626337, + "grad_norm": 1.1851018459560683, + "learning_rate": 1.614486725268426e-06, + "loss": 0.5713, + "step": 3133 + }, + { + "epoch": 0.7449488946993107, + "grad_norm": 1.3595050138856641, + "learning_rate": 1.6116545647908722e-06, + "loss": 0.5913, + "step": 3134 + }, + { + "epoch": 0.7451865937722842, + "grad_norm": 1.4003933668183797, + "learning_rate": 1.60882441325383e-06, + "loss": 0.5725, + "step": 3135 + }, + { + "epoch": 0.7454242928452579, + "grad_norm": 1.410282925566575, + "learning_rate": 1.6059962723352912e-06, + "loss": 0.6119, + "step": 3136 + }, + { + "epoch": 0.7456619919182316, + "grad_norm": 1.3697000442706664, + "learning_rate": 1.6031701437120512e-06, + "loss": 0.5603, + "step": 3137 + }, + { + "epoch": 0.7458996909912051, + "grad_norm": 1.402529298281416, + "learning_rate": 1.6003460290597162e-06, + "loss": 0.5598, + "step": 3138 + }, + { + "epoch": 0.7461373900641788, + "grad_norm": 1.6667737750475238, + "learning_rate": 1.5975239300526924e-06, + "loss": 0.5546, + "step": 3139 + }, + { + "epoch": 0.7463750891371523, + "grad_norm": 1.4177149831785683, + "learning_rate": 1.5947038483641964e-06, + "loss": 0.5468, + "step": 3140 + }, + { + "epoch": 0.746612788210126, + "grad_norm": 1.251320723594489, + "learning_rate": 1.5918857856662462e-06, + "loss": 0.5169, + "step": 3141 + }, + { + "epoch": 0.7468504872830996, + "grad_norm": 1.1986025330755654, + "learning_rate": 1.5890697436296648e-06, + "loss": 0.5344, + "step": 3142 + }, + { + "epoch": 0.7470881863560732, + "grad_norm": 1.2203635929467294, + "learning_rate": 1.5862557239240729e-06, + "loss": 0.5177, + "step": 3143 + }, + { + "epoch": 0.7473258854290469, + "grad_norm": 1.1776709638473777, + "learning_rate": 1.583443728217896e-06, + "loss": 0.5343, + "step": 3144 + }, + { + "epoch": 0.7475635845020204, + "grad_norm": 1.1572027489784404, + "learning_rate": 1.5806337581783593e-06, + "loss": 0.5568, + "step": 3145 + }, + { + "epoch": 0.7478012835749941, + "grad_norm": 1.2371178891497496, + "learning_rate": 1.577825815471487e-06, + "loss": 0.5257, + "step": 3146 + }, + { + "epoch": 0.7480389826479676, + "grad_norm": 1.6570595012106488, + "learning_rate": 1.575019901762097e-06, + "loss": 0.5409, + "step": 3147 + }, + { + "epoch": 0.7482766817209413, + "grad_norm": 1.2642728428905892, + "learning_rate": 1.5722160187138102e-06, + "loss": 0.5024, + "step": 3148 + }, + { + "epoch": 0.7485143807939149, + "grad_norm": 1.15097338764565, + "learning_rate": 1.5694141679890408e-06, + "loss": 0.5529, + "step": 3149 + }, + { + "epoch": 0.7487520798668885, + "grad_norm": 1.3020172279942965, + "learning_rate": 1.5666143512489996e-06, + "loss": 0.5436, + "step": 3150 + }, + { + "epoch": 0.7489897789398622, + "grad_norm": 1.346972457705643, + "learning_rate": 1.5638165701536866e-06, + "loss": 0.563, + "step": 3151 + }, + { + "epoch": 0.7492274780128357, + "grad_norm": 1.4635574779255518, + "learning_rate": 1.5610208263619002e-06, + "loss": 0.5326, + "step": 3152 + }, + { + "epoch": 0.7494651770858094, + "grad_norm": 1.3600309228946215, + "learning_rate": 1.5582271215312294e-06, + "loss": 0.6385, + "step": 3153 + }, + { + "epoch": 0.749702876158783, + "grad_norm": 1.1895646215294389, + "learning_rate": 1.5554354573180553e-06, + "loss": 0.5305, + "step": 3154 + }, + { + "epoch": 0.7499405752317566, + "grad_norm": 1.1571536866772305, + "learning_rate": 1.5526458353775447e-06, + "loss": 0.5435, + "step": 3155 + }, + { + "epoch": 0.7501782743047302, + "grad_norm": 1.101933194370475, + "learning_rate": 1.5498582573636578e-06, + "loss": 0.5997, + "step": 3156 + }, + { + "epoch": 0.7504159733777038, + "grad_norm": 1.2324533847354098, + "learning_rate": 1.5470727249291423e-06, + "loss": 0.5595, + "step": 3157 + }, + { + "epoch": 0.7506536724506775, + "grad_norm": 1.2413459928678106, + "learning_rate": 1.5442892397255337e-06, + "loss": 0.5833, + "step": 3158 + }, + { + "epoch": 0.750891371523651, + "grad_norm": 1.162492112345101, + "learning_rate": 1.54150780340315e-06, + "loss": 0.5598, + "step": 3159 + }, + { + "epoch": 0.7511290705966247, + "grad_norm": 1.1806071557785704, + "learning_rate": 1.5387284176110983e-06, + "loss": 0.4816, + "step": 3160 + }, + { + "epoch": 0.7513667696695983, + "grad_norm": 1.3434805371804108, + "learning_rate": 1.5359510839972685e-06, + "loss": 0.5735, + "step": 3161 + }, + { + "epoch": 0.7516044687425719, + "grad_norm": 1.2753366458715139, + "learning_rate": 1.5331758042083355e-06, + "loss": 0.5809, + "step": 3162 + }, + { + "epoch": 0.7518421678155455, + "grad_norm": 1.1355431362276578, + "learning_rate": 1.5304025798897521e-06, + "loss": 0.5226, + "step": 3163 + }, + { + "epoch": 0.7520798668885191, + "grad_norm": 1.463202712981851, + "learning_rate": 1.5276314126857566e-06, + "loss": 0.5797, + "step": 3164 + }, + { + "epoch": 0.7523175659614928, + "grad_norm": 1.7629024732365077, + "learning_rate": 1.5248623042393663e-06, + "loss": 0.5356, + "step": 3165 + }, + { + "epoch": 0.7525552650344663, + "grad_norm": 1.212540953990315, + "learning_rate": 1.5220952561923791e-06, + "loss": 0.5107, + "step": 3166 + }, + { + "epoch": 0.75279296410744, + "grad_norm": 1.3544105902794934, + "learning_rate": 1.5193302701853674e-06, + "loss": 0.5474, + "step": 3167 + }, + { + "epoch": 0.7530306631804136, + "grad_norm": 1.248018351612129, + "learning_rate": 1.5165673478576849e-06, + "loss": 0.5412, + "step": 3168 + }, + { + "epoch": 0.7532683622533872, + "grad_norm": 1.2282882752375752, + "learning_rate": 1.5138064908474603e-06, + "loss": 0.5436, + "step": 3169 + }, + { + "epoch": 0.7535060613263608, + "grad_norm": 1.5755598980215462, + "learning_rate": 1.5110477007916002e-06, + "loss": 0.5439, + "step": 3170 + }, + { + "epoch": 0.7537437603993344, + "grad_norm": 1.1507017451992578, + "learning_rate": 1.5082909793257795e-06, + "loss": 0.4913, + "step": 3171 + }, + { + "epoch": 0.7539814594723081, + "grad_norm": 1.288128178606889, + "learning_rate": 1.505536328084453e-06, + "loss": 0.5345, + "step": 3172 + }, + { + "epoch": 0.7542191585452817, + "grad_norm": 1.2236336356581805, + "learning_rate": 1.5027837487008456e-06, + "loss": 0.5802, + "step": 3173 + }, + { + "epoch": 0.7544568576182553, + "grad_norm": 1.5139159780165659, + "learning_rate": 1.5000332428069553e-06, + "loss": 0.5957, + "step": 3174 + }, + { + "epoch": 0.7546945566912289, + "grad_norm": 1.199143984558672, + "learning_rate": 1.4972848120335453e-06, + "loss": 0.5481, + "step": 3175 + }, + { + "epoch": 0.7549322557642025, + "grad_norm": 1.2801572495209717, + "learning_rate": 1.4945384580101557e-06, + "loss": 0.5587, + "step": 3176 + }, + { + "epoch": 0.7551699548371761, + "grad_norm": 1.3619116442251895, + "learning_rate": 1.4917941823650917e-06, + "loss": 0.5901, + "step": 3177 + }, + { + "epoch": 0.7554076539101497, + "grad_norm": 1.5210722661443663, + "learning_rate": 1.4890519867254271e-06, + "loss": 0.5713, + "step": 3178 + }, + { + "epoch": 0.7556453529831234, + "grad_norm": 1.2898482480467162, + "learning_rate": 1.486311872717e-06, + "loss": 0.5832, + "step": 3179 + }, + { + "epoch": 0.755883052056097, + "grad_norm": 1.1807130089380335, + "learning_rate": 1.4835738419644175e-06, + "loss": 0.5308, + "step": 3180 + }, + { + "epoch": 0.7561207511290706, + "grad_norm": 1.3036183801643373, + "learning_rate": 1.4808378960910502e-06, + "loss": 0.5571, + "step": 3181 + }, + { + "epoch": 0.7563584502020442, + "grad_norm": 1.4138269669334829, + "learning_rate": 1.478104036719034e-06, + "loss": 0.5691, + "step": 3182 + }, + { + "epoch": 0.7565961492750178, + "grad_norm": 1.390764406093566, + "learning_rate": 1.475372265469265e-06, + "loss": 0.5952, + "step": 3183 + }, + { + "epoch": 0.7568338483479914, + "grad_norm": 1.2518317017700447, + "learning_rate": 1.4726425839614022e-06, + "loss": 0.5834, + "step": 3184 + }, + { + "epoch": 0.757071547420965, + "grad_norm": 1.113849460131997, + "learning_rate": 1.469914993813867e-06, + "loss": 0.5574, + "step": 3185 + }, + { + "epoch": 0.7573092464939387, + "grad_norm": 1.1797963653641883, + "learning_rate": 1.4671894966438415e-06, + "loss": 0.5324, + "step": 3186 + }, + { + "epoch": 0.7575469455669123, + "grad_norm": 1.2170909618111858, + "learning_rate": 1.4644660940672628e-06, + "loss": 0.5357, + "step": 3187 + }, + { + "epoch": 0.7577846446398859, + "grad_norm": 1.2746676460883226, + "learning_rate": 1.4617447876988294e-06, + "loss": 0.5401, + "step": 3188 + }, + { + "epoch": 0.7580223437128595, + "grad_norm": 1.1499574339993994, + "learning_rate": 1.459025579151998e-06, + "loss": 0.5093, + "step": 3189 + }, + { + "epoch": 0.7582600427858331, + "grad_norm": 1.4386644981356043, + "learning_rate": 1.4563084700389768e-06, + "loss": 0.6204, + "step": 3190 + }, + { + "epoch": 0.7584977418588067, + "grad_norm": 1.2046708050717478, + "learning_rate": 1.453593461970733e-06, + "loss": 0.5152, + "step": 3191 + }, + { + "epoch": 0.7587354409317804, + "grad_norm": 1.16080625558366, + "learning_rate": 1.4508805565569883e-06, + "loss": 0.5415, + "step": 3192 + }, + { + "epoch": 0.758973140004754, + "grad_norm": 1.2458688933540232, + "learning_rate": 1.448169755406218e-06, + "loss": 0.552, + "step": 3193 + }, + { + "epoch": 0.7592108390777276, + "grad_norm": 1.5265968032179928, + "learning_rate": 1.4454610601256446e-06, + "loss": 0.5806, + "step": 3194 + }, + { + "epoch": 0.7594485381507012, + "grad_norm": 1.2363849406189422, + "learning_rate": 1.4427544723212489e-06, + "loss": 0.5853, + "step": 3195 + }, + { + "epoch": 0.7596862372236748, + "grad_norm": 1.2139413939350152, + "learning_rate": 1.440049993597758e-06, + "loss": 0.5326, + "step": 3196 + }, + { + "epoch": 0.7599239362966485, + "grad_norm": 1.2566271972243315, + "learning_rate": 1.4373476255586515e-06, + "loss": 0.5607, + "step": 3197 + }, + { + "epoch": 0.7601616353696221, + "grad_norm": 1.5742766506455856, + "learning_rate": 1.4346473698061536e-06, + "loss": 0.5546, + "step": 3198 + }, + { + "epoch": 0.7603993344425957, + "grad_norm": 1.4862389579513113, + "learning_rate": 1.4319492279412388e-06, + "loss": 0.577, + "step": 3199 + }, + { + "epoch": 0.7606370335155693, + "grad_norm": 1.3907664124231545, + "learning_rate": 1.429253201563629e-06, + "loss": 0.5534, + "step": 3200 + }, + { + "epoch": 0.7606370335155693, + "eval_loss": 0.5602092742919922, + "eval_runtime": 285.6371, + "eval_samples_per_second": 38.083, + "eval_steps_per_second": 1.19, + "step": 3200 + }, + { + "epoch": 0.7608747325885429, + "grad_norm": 1.219138152832234, + "learning_rate": 1.4265592922717908e-06, + "loss": 0.5548, + "step": 3201 + }, + { + "epoch": 0.7611124316615165, + "grad_norm": 1.3928724611555099, + "learning_rate": 1.423867501662934e-06, + "loss": 0.5598, + "step": 3202 + }, + { + "epoch": 0.7613501307344901, + "grad_norm": 1.2717439193374716, + "learning_rate": 1.4211778313330137e-06, + "loss": 0.593, + "step": 3203 + }, + { + "epoch": 0.7615878298074638, + "grad_norm": 1.2184805915086607, + "learning_rate": 1.4184902828767288e-06, + "loss": 0.5311, + "step": 3204 + }, + { + "epoch": 0.7618255288804374, + "grad_norm": 1.1937295242326595, + "learning_rate": 1.4158048578875211e-06, + "loss": 0.5165, + "step": 3205 + }, + { + "epoch": 0.762063227953411, + "grad_norm": 1.186379420550643, + "learning_rate": 1.4131215579575674e-06, + "loss": 0.5283, + "step": 3206 + }, + { + "epoch": 0.7623009270263846, + "grad_norm": 1.2526760965167132, + "learning_rate": 1.410440384677791e-06, + "loss": 0.591, + "step": 3207 + }, + { + "epoch": 0.7625386260993582, + "grad_norm": 1.2901289280981914, + "learning_rate": 1.407761339637852e-06, + "loss": 0.5548, + "step": 3208 + }, + { + "epoch": 0.7627763251723318, + "grad_norm": 1.2903591656425093, + "learning_rate": 1.4050844244261496e-06, + "loss": 0.5364, + "step": 3209 + }, + { + "epoch": 0.7630140242453054, + "grad_norm": 1.525881575004847, + "learning_rate": 1.4024096406298176e-06, + "loss": 0.5877, + "step": 3210 + }, + { + "epoch": 0.7632517233182791, + "grad_norm": 1.2671949775966875, + "learning_rate": 1.399736989834728e-06, + "loss": 0.5315, + "step": 3211 + }, + { + "epoch": 0.7634894223912527, + "grad_norm": 1.364078795490182, + "learning_rate": 1.3970664736254885e-06, + "loss": 0.5514, + "step": 3212 + }, + { + "epoch": 0.7637271214642263, + "grad_norm": 1.0691008046523858, + "learning_rate": 1.3943980935854418e-06, + "loss": 0.5574, + "step": 3213 + }, + { + "epoch": 0.7639648205371999, + "grad_norm": 1.2297194471395776, + "learning_rate": 1.391731851296661e-06, + "loss": 0.542, + "step": 3214 + }, + { + "epoch": 0.7642025196101735, + "grad_norm": 1.4062766055363707, + "learning_rate": 1.389067748339954e-06, + "loss": 0.5456, + "step": 3215 + }, + { + "epoch": 0.7644402186831472, + "grad_norm": 1.7794633378641351, + "learning_rate": 1.3864057862948604e-06, + "loss": 0.5428, + "step": 3216 + }, + { + "epoch": 0.7646779177561207, + "grad_norm": 1.2253484145613849, + "learning_rate": 1.383745966739652e-06, + "loss": 0.5913, + "step": 3217 + }, + { + "epoch": 0.7649156168290944, + "grad_norm": 1.1739372013977687, + "learning_rate": 1.3810882912513246e-06, + "loss": 0.5818, + "step": 3218 + }, + { + "epoch": 0.765153315902068, + "grad_norm": 1.2251046940406378, + "learning_rate": 1.3784327614056087e-06, + "loss": 0.5388, + "step": 3219 + }, + { + "epoch": 0.7653910149750416, + "grad_norm": 1.4123407488052049, + "learning_rate": 1.3757793787769596e-06, + "loss": 0.595, + "step": 3220 + }, + { + "epoch": 0.7656287140480152, + "grad_norm": 1.3823285844949023, + "learning_rate": 1.373128144938563e-06, + "loss": 0.5429, + "step": 3221 + }, + { + "epoch": 0.7658664131209888, + "grad_norm": 1.1951133330372088, + "learning_rate": 1.3704790614623248e-06, + "loss": 0.5657, + "step": 3222 + }, + { + "epoch": 0.7661041121939625, + "grad_norm": 1.3640226010996306, + "learning_rate": 1.3678321299188802e-06, + "loss": 0.5914, + "step": 3223 + }, + { + "epoch": 0.766341811266936, + "grad_norm": 1.1268262588715743, + "learning_rate": 1.3651873518775876e-06, + "loss": 0.5883, + "step": 3224 + }, + { + "epoch": 0.7665795103399097, + "grad_norm": 1.4437470476746361, + "learning_rate": 1.3625447289065302e-06, + "loss": 0.5701, + "step": 3225 + }, + { + "epoch": 0.7668172094128833, + "grad_norm": 1.2341217714302413, + "learning_rate": 1.3599042625725084e-06, + "loss": 0.5757, + "step": 3226 + }, + { + "epoch": 0.7670549084858569, + "grad_norm": 1.1855475297302647, + "learning_rate": 1.3572659544410493e-06, + "loss": 0.5215, + "step": 3227 + }, + { + "epoch": 0.7672926075588306, + "grad_norm": 1.4584722397041083, + "learning_rate": 1.354629806076398e-06, + "loss": 0.5659, + "step": 3228 + }, + { + "epoch": 0.7675303066318041, + "grad_norm": 1.9027683264933564, + "learning_rate": 1.351995819041521e-06, + "loss": 0.5556, + "step": 3229 + }, + { + "epoch": 0.7677680057047778, + "grad_norm": 1.1936078927476212, + "learning_rate": 1.349363994898099e-06, + "loss": 0.5189, + "step": 3230 + }, + { + "epoch": 0.7680057047777513, + "grad_norm": 1.2908197591139314, + "learning_rate": 1.3467343352065349e-06, + "loss": 0.5503, + "step": 3231 + }, + { + "epoch": 0.768243403850725, + "grad_norm": 1.1327025430418562, + "learning_rate": 1.3441068415259462e-06, + "loss": 0.5749, + "step": 3232 + }, + { + "epoch": 0.7684811029236986, + "grad_norm": 2.5454539508894376, + "learning_rate": 1.3414815154141682e-06, + "loss": 0.6023, + "step": 3233 + }, + { + "epoch": 0.7687188019966722, + "grad_norm": 1.1122369797359912, + "learning_rate": 1.3388583584277459e-06, + "loss": 0.5428, + "step": 3234 + }, + { + "epoch": 0.7689565010696459, + "grad_norm": 1.2661916485488138, + "learning_rate": 1.336237372121944e-06, + "loss": 0.5373, + "step": 3235 + }, + { + "epoch": 0.7691942001426194, + "grad_norm": 1.2645830990629472, + "learning_rate": 1.3336185580507377e-06, + "loss": 0.5508, + "step": 3236 + }, + { + "epoch": 0.7694318992155931, + "grad_norm": 1.2095671201875353, + "learning_rate": 1.3310019177668154e-06, + "loss": 0.5417, + "step": 3237 + }, + { + "epoch": 0.7696695982885666, + "grad_norm": 1.4959950671065916, + "learning_rate": 1.3283874528215735e-06, + "loss": 0.5868, + "step": 3238 + }, + { + "epoch": 0.7699072973615403, + "grad_norm": 1.294941910477183, + "learning_rate": 1.3257751647651223e-06, + "loss": 0.5345, + "step": 3239 + }, + { + "epoch": 0.770144996434514, + "grad_norm": 1.2955330289704197, + "learning_rate": 1.32316505514628e-06, + "loss": 0.5247, + "step": 3240 + }, + { + "epoch": 0.7703826955074875, + "grad_norm": 1.1160879536609105, + "learning_rate": 1.320557125512575e-06, + "loss": 0.5364, + "step": 3241 + }, + { + "epoch": 0.7706203945804612, + "grad_norm": 1.2371248518660767, + "learning_rate": 1.3179513774102376e-06, + "loss": 0.5822, + "step": 3242 + }, + { + "epoch": 0.7708580936534347, + "grad_norm": 1.2478695639626949, + "learning_rate": 1.3153478123842111e-06, + "loss": 0.5556, + "step": 3243 + }, + { + "epoch": 0.7710957927264084, + "grad_norm": 1.3120178202021928, + "learning_rate": 1.3127464319781413e-06, + "loss": 0.5476, + "step": 3244 + }, + { + "epoch": 0.7713334917993819, + "grad_norm": 1.2378523511363957, + "learning_rate": 1.3101472377343815e-06, + "loss": 0.578, + "step": 3245 + }, + { + "epoch": 0.7715711908723556, + "grad_norm": 1.2291248004710558, + "learning_rate": 1.307550231193983e-06, + "loss": 0.5305, + "step": 3246 + }, + { + "epoch": 0.7718088899453293, + "grad_norm": 1.460548212423444, + "learning_rate": 1.3049554138967052e-06, + "loss": 0.5777, + "step": 3247 + }, + { + "epoch": 0.7720465890183028, + "grad_norm": 1.3170448068441007, + "learning_rate": 1.3023627873810086e-06, + "loss": 0.5301, + "step": 3248 + }, + { + "epoch": 0.7722842880912765, + "grad_norm": 1.332606296312683, + "learning_rate": 1.299772353184055e-06, + "loss": 0.5365, + "step": 3249 + }, + { + "epoch": 0.77252198716425, + "grad_norm": 1.2336229158494263, + "learning_rate": 1.2971841128417034e-06, + "loss": 0.5897, + "step": 3250 + }, + { + "epoch": 0.7727596862372237, + "grad_norm": 1.4617603443071037, + "learning_rate": 1.2945980678885134e-06, + "loss": 0.5601, + "step": 3251 + }, + { + "epoch": 0.7729973853101972, + "grad_norm": 1.3134891389006467, + "learning_rate": 1.2920142198577484e-06, + "loss": 0.5331, + "step": 3252 + }, + { + "epoch": 0.7732350843831709, + "grad_norm": 1.3591237077102212, + "learning_rate": 1.289432570281361e-06, + "loss": 0.618, + "step": 3253 + }, + { + "epoch": 0.7734727834561446, + "grad_norm": 1.0993571895044871, + "learning_rate": 1.2868531206900048e-06, + "loss": 0.5511, + "step": 3254 + }, + { + "epoch": 0.7737104825291181, + "grad_norm": 1.3140542580526293, + "learning_rate": 1.2842758726130283e-06, + "loss": 0.5395, + "step": 3255 + }, + { + "epoch": 0.7739481816020918, + "grad_norm": 1.1614385781537606, + "learning_rate": 1.281700827578476e-06, + "loss": 0.553, + "step": 3256 + }, + { + "epoch": 0.7741858806750653, + "grad_norm": 1.0832642842383498, + "learning_rate": 1.2791279871130824e-06, + "loss": 0.5585, + "step": 3257 + }, + { + "epoch": 0.774423579748039, + "grad_norm": 1.2857776852472358, + "learning_rate": 1.2765573527422785e-06, + "loss": 0.5477, + "step": 3258 + }, + { + "epoch": 0.7746612788210125, + "grad_norm": 1.0469253437264956, + "learning_rate": 1.2739889259901866e-06, + "loss": 0.545, + "step": 3259 + }, + { + "epoch": 0.7748989778939862, + "grad_norm": 1.087361518735006, + "learning_rate": 1.2714227083796205e-06, + "loss": 0.597, + "step": 3260 + }, + { + "epoch": 0.7751366769669599, + "grad_norm": 1.2409143001373122, + "learning_rate": 1.2688587014320809e-06, + "loss": 0.6171, + "step": 3261 + }, + { + "epoch": 0.7753743760399334, + "grad_norm": 1.1574154018074405, + "learning_rate": 1.266296906667762e-06, + "loss": 0.572, + "step": 3262 + }, + { + "epoch": 0.7756120751129071, + "grad_norm": 1.0943775777391447, + "learning_rate": 1.2637373256055445e-06, + "loss": 0.5464, + "step": 3263 + }, + { + "epoch": 0.7758497741858806, + "grad_norm": 1.6196073804015003, + "learning_rate": 1.2611799597629987e-06, + "loss": 0.5337, + "step": 3264 + }, + { + "epoch": 0.7760874732588543, + "grad_norm": 1.3066008013515324, + "learning_rate": 1.258624810656376e-06, + "loss": 0.5487, + "step": 3265 + }, + { + "epoch": 0.7763251723318279, + "grad_norm": 1.249991162247289, + "learning_rate": 1.2560718798006188e-06, + "loss": 0.569, + "step": 3266 + }, + { + "epoch": 0.7765628714048015, + "grad_norm": 1.1984171024430252, + "learning_rate": 1.2535211687093535e-06, + "loss": 0.5846, + "step": 3267 + }, + { + "epoch": 0.7768005704777752, + "grad_norm": 1.4118613980855912, + "learning_rate": 1.2509726788948894e-06, + "loss": 0.5881, + "step": 3268 + }, + { + "epoch": 0.7770382695507487, + "grad_norm": 1.1781246540518657, + "learning_rate": 1.2484264118682176e-06, + "loss": 0.5762, + "step": 3269 + }, + { + "epoch": 0.7772759686237224, + "grad_norm": 1.1875125793736385, + "learning_rate": 1.2458823691390137e-06, + "loss": 0.5153, + "step": 3270 + }, + { + "epoch": 0.7775136676966959, + "grad_norm": 1.417417447800564, + "learning_rate": 1.2433405522156334e-06, + "loss": 0.5579, + "step": 3271 + }, + { + "epoch": 0.7777513667696696, + "grad_norm": 1.2353102228151616, + "learning_rate": 1.2408009626051137e-06, + "loss": 0.5787, + "step": 3272 + }, + { + "epoch": 0.7779890658426433, + "grad_norm": 2.4249319811158023, + "learning_rate": 1.2382636018131678e-06, + "loss": 0.5632, + "step": 3273 + }, + { + "epoch": 0.7782267649156168, + "grad_norm": 2.2447563104650574, + "learning_rate": 1.235728471344192e-06, + "loss": 0.6026, + "step": 3274 + }, + { + "epoch": 0.7784644639885905, + "grad_norm": 1.6279504816451553, + "learning_rate": 1.2331955727012562e-06, + "loss": 0.5538, + "step": 3275 + }, + { + "epoch": 0.778702163061564, + "grad_norm": 1.1431798284499535, + "learning_rate": 1.2306649073861121e-06, + "loss": 0.5275, + "step": 3276 + }, + { + "epoch": 0.7789398621345377, + "grad_norm": 1.815463414204192, + "learning_rate": 1.2281364768991804e-06, + "loss": 0.5733, + "step": 3277 + }, + { + "epoch": 0.7791775612075112, + "grad_norm": 1.2111881012766494, + "learning_rate": 1.2256102827395622e-06, + "loss": 0.5024, + "step": 3278 + }, + { + "epoch": 0.7794152602804849, + "grad_norm": 1.2793887149963254, + "learning_rate": 1.2230863264050308e-06, + "loss": 0.5574, + "step": 3279 + }, + { + "epoch": 0.7796529593534586, + "grad_norm": 1.5048471171853361, + "learning_rate": 1.2205646093920342e-06, + "loss": 0.5493, + "step": 3280 + }, + { + "epoch": 0.7798906584264321, + "grad_norm": 1.431220125648615, + "learning_rate": 1.2180451331956883e-06, + "loss": 0.513, + "step": 3281 + }, + { + "epoch": 0.7801283574994058, + "grad_norm": 1.219508278811579, + "learning_rate": 1.2155278993097853e-06, + "loss": 0.537, + "step": 3282 + }, + { + "epoch": 0.7803660565723793, + "grad_norm": 1.2973149599187446, + "learning_rate": 1.213012909226786e-06, + "loss": 0.5481, + "step": 3283 + }, + { + "epoch": 0.780603755645353, + "grad_norm": 1.3434868238839321, + "learning_rate": 1.2105001644378223e-06, + "loss": 0.5224, + "step": 3284 + }, + { + "epoch": 0.7808414547183266, + "grad_norm": 1.417877126053815, + "learning_rate": 1.2079896664326906e-06, + "loss": 0.6118, + "step": 3285 + }, + { + "epoch": 0.7810791537913002, + "grad_norm": 1.1501232415033957, + "learning_rate": 1.2054814166998596e-06, + "loss": 0.5211, + "step": 3286 + }, + { + "epoch": 0.7813168528642739, + "grad_norm": 1.3419499864632887, + "learning_rate": 1.202975416726464e-06, + "loss": 0.5709, + "step": 3287 + }, + { + "epoch": 0.7815545519372474, + "grad_norm": 1.239523255296169, + "learning_rate": 1.2004716679983052e-06, + "loss": 0.5664, + "step": 3288 + }, + { + "epoch": 0.7817922510102211, + "grad_norm": 1.1969580520156622, + "learning_rate": 1.1979701719998454e-06, + "loss": 0.5631, + "step": 3289 + }, + { + "epoch": 0.7820299500831946, + "grad_norm": 1.4141435019246795, + "learning_rate": 1.1954709302142176e-06, + "loss": 0.5802, + "step": 3290 + }, + { + "epoch": 0.7822676491561683, + "grad_norm": 1.1703648898380448, + "learning_rate": 1.1929739441232136e-06, + "loss": 0.5362, + "step": 3291 + }, + { + "epoch": 0.7825053482291419, + "grad_norm": 1.3198674135622648, + "learning_rate": 1.1904792152072914e-06, + "loss": 0.5139, + "step": 3292 + }, + { + "epoch": 0.7827430473021155, + "grad_norm": 1.1895035603966657, + "learning_rate": 1.1879867449455663e-06, + "loss": 0.5605, + "step": 3293 + }, + { + "epoch": 0.7829807463750892, + "grad_norm": 1.125619566635305, + "learning_rate": 1.1854965348158177e-06, + "loss": 0.5526, + "step": 3294 + }, + { + "epoch": 0.7832184454480627, + "grad_norm": 1.230835167388697, + "learning_rate": 1.1830085862944851e-06, + "loss": 0.5324, + "step": 3295 + }, + { + "epoch": 0.7834561445210364, + "grad_norm": 1.335068776398367, + "learning_rate": 1.1805229008566665e-06, + "loss": 0.5976, + "step": 3296 + }, + { + "epoch": 0.78369384359401, + "grad_norm": 1.2625406629649498, + "learning_rate": 1.1780394799761163e-06, + "loss": 0.5302, + "step": 3297 + }, + { + "epoch": 0.7839315426669836, + "grad_norm": 1.1754766751380443, + "learning_rate": 1.1755583251252484e-06, + "loss": 0.5571, + "step": 3298 + }, + { + "epoch": 0.7841692417399572, + "grad_norm": 1.2219957564414223, + "learning_rate": 1.1730794377751327e-06, + "loss": 0.5084, + "step": 3299 + }, + { + "epoch": 0.7844069408129308, + "grad_norm": 1.4304118458703918, + "learning_rate": 1.1706028193954954e-06, + "loss": 0.5514, + "step": 3300 + }, + { + "epoch": 0.7844069408129308, + "eval_loss": 0.5583809018135071, + "eval_runtime": 279.0869, + "eval_samples_per_second": 38.977, + "eval_steps_per_second": 1.218, + "step": 3300 + }, + { + "epoch": 0.7846446398859045, + "grad_norm": 1.2428410947570614, + "learning_rate": 1.1681284714547147e-06, + "loss": 0.6017, + "step": 3301 + }, + { + "epoch": 0.784882338958878, + "grad_norm": 1.2589505258204103, + "learning_rate": 1.1656563954198258e-06, + "loss": 0.5615, + "step": 3302 + }, + { + "epoch": 0.7851200380318517, + "grad_norm": 1.1656754214971796, + "learning_rate": 1.163186592756515e-06, + "loss": 0.5545, + "step": 3303 + }, + { + "epoch": 0.7853577371048253, + "grad_norm": 1.2919164800786194, + "learning_rate": 1.1607190649291239e-06, + "loss": 0.5391, + "step": 3304 + }, + { + "epoch": 0.7855954361777989, + "grad_norm": 1.2399325438138868, + "learning_rate": 1.1582538134006388e-06, + "loss": 0.5402, + "step": 3305 + }, + { + "epoch": 0.7858331352507725, + "grad_norm": 1.5322909734565882, + "learning_rate": 1.1557908396327028e-06, + "loss": 0.5297, + "step": 3306 + }, + { + "epoch": 0.7860708343237461, + "grad_norm": 1.2154463748011397, + "learning_rate": 1.1533301450856054e-06, + "loss": 0.5892, + "step": 3307 + }, + { + "epoch": 0.7863085333967198, + "grad_norm": 1.6361256937913298, + "learning_rate": 1.150871731218287e-06, + "loss": 0.4746, + "step": 3308 + }, + { + "epoch": 0.7865462324696934, + "grad_norm": 1.2200135392510068, + "learning_rate": 1.1484155994883323e-06, + "loss": 0.5117, + "step": 3309 + }, + { + "epoch": 0.786783931542667, + "grad_norm": 1.2939841773576932, + "learning_rate": 1.1459617513519756e-06, + "loss": 0.5852, + "step": 3310 + }, + { + "epoch": 0.7870216306156406, + "grad_norm": 1.2586720573957768, + "learning_rate": 1.1435101882640964e-06, + "loss": 0.5642, + "step": 3311 + }, + { + "epoch": 0.7872593296886142, + "grad_norm": 1.4225125208118112, + "learning_rate": 1.1410609116782218e-06, + "loss": 0.5387, + "step": 3312 + }, + { + "epoch": 0.7874970287615878, + "grad_norm": 1.1716574463957141, + "learning_rate": 1.1386139230465176e-06, + "loss": 0.4832, + "step": 3313 + }, + { + "epoch": 0.7877347278345614, + "grad_norm": 1.1993973820154966, + "learning_rate": 1.136169223819798e-06, + "loss": 0.5077, + "step": 3314 + }, + { + "epoch": 0.7879724269075351, + "grad_norm": 1.3835243297855575, + "learning_rate": 1.1337268154475184e-06, + "loss": 0.5591, + "step": 3315 + }, + { + "epoch": 0.7882101259805087, + "grad_norm": 1.3247943553990256, + "learning_rate": 1.1312866993777771e-06, + "loss": 0.5369, + "step": 3316 + }, + { + "epoch": 0.7884478250534823, + "grad_norm": 1.4881533845840593, + "learning_rate": 1.1288488770573097e-06, + "loss": 0.5427, + "step": 3317 + }, + { + "epoch": 0.7886855241264559, + "grad_norm": 1.163845824849692, + "learning_rate": 1.126413349931495e-06, + "loss": 0.5088, + "step": 3318 + }, + { + "epoch": 0.7889232231994295, + "grad_norm": 1.4153929131524043, + "learning_rate": 1.1239801194443507e-06, + "loss": 0.5462, + "step": 3319 + }, + { + "epoch": 0.7891609222724031, + "grad_norm": 1.1931361602643218, + "learning_rate": 1.1215491870385336e-06, + "loss": 0.5153, + "step": 3320 + }, + { + "epoch": 0.7893986213453767, + "grad_norm": 1.33085253888579, + "learning_rate": 1.119120554155334e-06, + "loss": 0.6082, + "step": 3321 + }, + { + "epoch": 0.7896363204183504, + "grad_norm": 1.321555295328739, + "learning_rate": 1.1166942222346828e-06, + "loss": 0.5918, + "step": 3322 + }, + { + "epoch": 0.789874019491324, + "grad_norm": 1.3287352385222078, + "learning_rate": 1.1142701927151456e-06, + "loss": 0.5802, + "step": 3323 + }, + { + "epoch": 0.7901117185642976, + "grad_norm": 1.328174819178582, + "learning_rate": 1.1118484670339235e-06, + "loss": 0.5239, + "step": 3324 + }, + { + "epoch": 0.7903494176372712, + "grad_norm": 1.4092490801097457, + "learning_rate": 1.1094290466268493e-06, + "loss": 0.5802, + "step": 3325 + }, + { + "epoch": 0.7905871167102448, + "grad_norm": 1.2286450770276462, + "learning_rate": 1.10701193292839e-06, + "loss": 0.5406, + "step": 3326 + }, + { + "epoch": 0.7908248157832184, + "grad_norm": 1.215890241381548, + "learning_rate": 1.1045971273716476e-06, + "loss": 0.5503, + "step": 3327 + }, + { + "epoch": 0.7910625148561921, + "grad_norm": 1.5031859179495026, + "learning_rate": 1.1021846313883539e-06, + "loss": 0.5534, + "step": 3328 + }, + { + "epoch": 0.7913002139291657, + "grad_norm": 2.4565845515373406, + "learning_rate": 1.0997744464088684e-06, + "loss": 0.5741, + "step": 3329 + }, + { + "epoch": 0.7915379130021393, + "grad_norm": 1.258559330654481, + "learning_rate": 1.0973665738621847e-06, + "loss": 0.5371, + "step": 3330 + }, + { + "epoch": 0.7917756120751129, + "grad_norm": 1.224596985223719, + "learning_rate": 1.0949610151759233e-06, + "loss": 0.5255, + "step": 3331 + }, + { + "epoch": 0.7920133111480865, + "grad_norm": 1.3257122980371743, + "learning_rate": 1.0925577717763342e-06, + "loss": 0.513, + "step": 3332 + }, + { + "epoch": 0.7922510102210601, + "grad_norm": 1.27370371325229, + "learning_rate": 1.0901568450882915e-06, + "loss": 0.5464, + "step": 3333 + }, + { + "epoch": 0.7924887092940337, + "grad_norm": 1.2258377529507078, + "learning_rate": 1.0877582365352995e-06, + "loss": 0.5474, + "step": 3334 + }, + { + "epoch": 0.7927264083670074, + "grad_norm": 1.705637674985423, + "learning_rate": 1.085361947539486e-06, + "loss": 0.5624, + "step": 3335 + }, + { + "epoch": 0.792964107439981, + "grad_norm": 1.5242659598426687, + "learning_rate": 1.082967979521606e-06, + "loss": 0.5212, + "step": 3336 + }, + { + "epoch": 0.7932018065129546, + "grad_norm": 1.2932575644720636, + "learning_rate": 1.0805763339010329e-06, + "loss": 0.5473, + "step": 3337 + }, + { + "epoch": 0.7934395055859282, + "grad_norm": 1.3386481083783663, + "learning_rate": 1.0781870120957683e-06, + "loss": 0.5441, + "step": 3338 + }, + { + "epoch": 0.7936772046589018, + "grad_norm": 1.3316931857832888, + "learning_rate": 1.0758000155224341e-06, + "loss": 0.5568, + "step": 3339 + }, + { + "epoch": 0.7939149037318755, + "grad_norm": 1.2437764089996994, + "learning_rate": 1.0734153455962765e-06, + "loss": 0.5477, + "step": 3340 + }, + { + "epoch": 0.794152602804849, + "grad_norm": 1.8663762610072272, + "learning_rate": 1.0710330037311556e-06, + "loss": 0.5132, + "step": 3341 + }, + { + "epoch": 0.7943903018778227, + "grad_norm": 1.7145564787981584, + "learning_rate": 1.0686529913395572e-06, + "loss": 0.5661, + "step": 3342 + }, + { + "epoch": 0.7946280009507963, + "grad_norm": 1.3189545996468512, + "learning_rate": 1.066275309832584e-06, + "loss": 0.5937, + "step": 3343 + }, + { + "epoch": 0.7948657000237699, + "grad_norm": 1.1263127319506607, + "learning_rate": 1.0638999606199573e-06, + "loss": 0.4722, + "step": 3344 + }, + { + "epoch": 0.7951033990967435, + "grad_norm": 1.6548120873674679, + "learning_rate": 1.0615269451100136e-06, + "loss": 0.5402, + "step": 3345 + }, + { + "epoch": 0.7953410981697171, + "grad_norm": 1.293793317469771, + "learning_rate": 1.059156264709707e-06, + "loss": 0.5939, + "step": 3346 + }, + { + "epoch": 0.7955787972426908, + "grad_norm": 1.1659518097529085, + "learning_rate": 1.0567879208246084e-06, + "loss": 0.5672, + "step": 3347 + }, + { + "epoch": 0.7958164963156644, + "grad_norm": 1.2056060239270292, + "learning_rate": 1.0544219148589036e-06, + "loss": 0.518, + "step": 3348 + }, + { + "epoch": 0.796054195388638, + "grad_norm": 1.1096898573675646, + "learning_rate": 1.0520582482153874e-06, + "loss": 0.5391, + "step": 3349 + }, + { + "epoch": 0.7962918944616116, + "grad_norm": 1.286162764105734, + "learning_rate": 1.0496969222954728e-06, + "loss": 0.5648, + "step": 3350 + }, + { + "epoch": 0.7965295935345852, + "grad_norm": 1.2216041224494183, + "learning_rate": 1.0473379384991833e-06, + "loss": 0.5747, + "step": 3351 + }, + { + "epoch": 0.7967672926075589, + "grad_norm": 1.1827293196154742, + "learning_rate": 1.0449812982251556e-06, + "loss": 0.5062, + "step": 3352 + }, + { + "epoch": 0.7970049916805324, + "grad_norm": 1.3490204143346538, + "learning_rate": 1.042627002870632e-06, + "loss": 0.5656, + "step": 3353 + }, + { + "epoch": 0.7972426907535061, + "grad_norm": 1.2537864427508156, + "learning_rate": 1.0402750538314694e-06, + "loss": 0.5462, + "step": 3354 + }, + { + "epoch": 0.7974803898264797, + "grad_norm": 1.844553104170143, + "learning_rate": 1.037925452502131e-06, + "loss": 0.5457, + "step": 3355 + }, + { + "epoch": 0.7977180888994533, + "grad_norm": 1.6458679839277526, + "learning_rate": 1.0355782002756908e-06, + "loss": 0.562, + "step": 3356 + }, + { + "epoch": 0.7979557879724269, + "grad_norm": 1.2422655683112815, + "learning_rate": 1.0332332985438248e-06, + "loss": 0.5441, + "step": 3357 + }, + { + "epoch": 0.7981934870454005, + "grad_norm": 1.6289602473813578, + "learning_rate": 1.0308907486968212e-06, + "loss": 0.515, + "step": 3358 + }, + { + "epoch": 0.7984311861183742, + "grad_norm": 1.2818417642047062, + "learning_rate": 1.02855055212357e-06, + "loss": 0.5963, + "step": 3359 + }, + { + "epoch": 0.7986688851913477, + "grad_norm": 1.1335626137649826, + "learning_rate": 1.0262127102115687e-06, + "loss": 0.5369, + "step": 3360 + }, + { + "epoch": 0.7989065842643214, + "grad_norm": 1.3036640001997049, + "learning_rate": 1.0238772243469153e-06, + "loss": 0.5128, + "step": 3361 + }, + { + "epoch": 0.799144283337295, + "grad_norm": 1.14165659239785, + "learning_rate": 1.0215440959143137e-06, + "loss": 0.5483, + "step": 3362 + }, + { + "epoch": 0.7993819824102686, + "grad_norm": 1.3044538078172103, + "learning_rate": 1.0192133262970688e-06, + "loss": 0.4991, + "step": 3363 + }, + { + "epoch": 0.7996196814832423, + "grad_norm": 1.2953862988907627, + "learning_rate": 1.0168849168770894e-06, + "loss": 0.5417, + "step": 3364 + }, + { + "epoch": 0.7998573805562158, + "grad_norm": 1.2641053736148495, + "learning_rate": 1.0145588690348795e-06, + "loss": 0.6109, + "step": 3365 + }, + { + "epoch": 0.8000950796291895, + "grad_norm": 1.2121409165516872, + "learning_rate": 1.0122351841495487e-06, + "loss": 0.5749, + "step": 3366 + }, + { + "epoch": 0.800332778702163, + "grad_norm": 1.442939018036069, + "learning_rate": 1.0099138635988026e-06, + "loss": 0.5788, + "step": 3367 + }, + { + "epoch": 0.8005704777751367, + "grad_norm": 1.2842410228856902, + "learning_rate": 1.007594908758947e-06, + "loss": 0.5098, + "step": 3368 + }, + { + "epoch": 0.8008081768481103, + "grad_norm": 1.3234560330757925, + "learning_rate": 1.0052783210048812e-06, + "loss": 0.5531, + "step": 3369 + }, + { + "epoch": 0.8010458759210839, + "grad_norm": 1.2897414854742466, + "learning_rate": 1.0029641017101049e-06, + "loss": 0.6022, + "step": 3370 + }, + { + "epoch": 0.8012835749940576, + "grad_norm": 1.1154109679933764, + "learning_rate": 1.0006522522467121e-06, + "loss": 0.5314, + "step": 3371 + }, + { + "epoch": 0.8015212740670311, + "grad_norm": 1.264708869633613, + "learning_rate": 9.983427739853935e-07, + "loss": 0.5418, + "step": 3372 + }, + { + "epoch": 0.8017589731400048, + "grad_norm": 1.33470795737269, + "learning_rate": 9.960356682954293e-07, + "loss": 0.5373, + "step": 3373 + }, + { + "epoch": 0.8019966722129783, + "grad_norm": 1.8059523818536225, + "learning_rate": 9.937309365446973e-07, + "loss": 0.5016, + "step": 3374 + }, + { + "epoch": 0.802234371285952, + "grad_norm": 1.3931853777536765, + "learning_rate": 9.91428580099667e-07, + "loss": 0.5027, + "step": 3375 + }, + { + "epoch": 0.8024720703589256, + "grad_norm": 1.1675410153087276, + "learning_rate": 9.891286003253997e-07, + "loss": 0.5662, + "step": 3376 + }, + { + "epoch": 0.8027097694318992, + "grad_norm": 1.1731537196075799, + "learning_rate": 9.868309985855446e-07, + "loss": 0.5482, + "step": 3377 + }, + { + "epoch": 0.8029474685048729, + "grad_norm": 1.3732719956749257, + "learning_rate": 9.845357762423453e-07, + "loss": 0.5849, + "step": 3378 + }, + { + "epoch": 0.8031851675778464, + "grad_norm": 1.4309902395991705, + "learning_rate": 9.822429346566314e-07, + "loss": 0.5334, + "step": 3379 + }, + { + "epoch": 0.8034228666508201, + "grad_norm": 1.5020868408399217, + "learning_rate": 9.799524751878236e-07, + "loss": 0.5656, + "step": 3380 + }, + { + "epoch": 0.8036605657237936, + "grad_norm": 1.2448973311574332, + "learning_rate": 9.776643991939283e-07, + "loss": 0.5252, + "step": 3381 + }, + { + "epoch": 0.8038982647967673, + "grad_norm": 1.2283515749760492, + "learning_rate": 9.753787080315385e-07, + "loss": 0.5465, + "step": 3382 + }, + { + "epoch": 0.804135963869741, + "grad_norm": 1.1620124700769616, + "learning_rate": 9.73095403055837e-07, + "loss": 0.5064, + "step": 3383 + }, + { + "epoch": 0.8043736629427145, + "grad_norm": 2.0225445891011504, + "learning_rate": 9.708144856205853e-07, + "loss": 0.5573, + "step": 3384 + }, + { + "epoch": 0.8046113620156882, + "grad_norm": 1.6873526664510774, + "learning_rate": 9.685359570781344e-07, + "loss": 0.547, + "step": 3385 + }, + { + "epoch": 0.8048490610886617, + "grad_norm": 1.1910744805891567, + "learning_rate": 9.66259818779418e-07, + "loss": 0.5602, + "step": 3386 + }, + { + "epoch": 0.8050867601616354, + "grad_norm": 1.1213874444238934, + "learning_rate": 9.639860720739524e-07, + "loss": 0.5427, + "step": 3387 + }, + { + "epoch": 0.8053244592346089, + "grad_norm": 1.2285406917382802, + "learning_rate": 9.617147183098341e-07, + "loss": 0.5753, + "step": 3388 + }, + { + "epoch": 0.8055621583075826, + "grad_norm": 1.3058982737239506, + "learning_rate": 9.59445758833743e-07, + "loss": 0.5777, + "step": 3389 + }, + { + "epoch": 0.8057998573805563, + "grad_norm": 1.4134755078091705, + "learning_rate": 9.571791949909392e-07, + "loss": 0.5444, + "step": 3390 + }, + { + "epoch": 0.8060375564535298, + "grad_norm": 1.1809846396141799, + "learning_rate": 9.549150281252633e-07, + "loss": 0.5864, + "step": 3391 + }, + { + "epoch": 0.8062752555265035, + "grad_norm": 1.517006251878074, + "learning_rate": 9.526532595791305e-07, + "loss": 0.568, + "step": 3392 + }, + { + "epoch": 0.806512954599477, + "grad_norm": 1.251258950308915, + "learning_rate": 9.503938906935389e-07, + "loss": 0.5699, + "step": 3393 + }, + { + "epoch": 0.8067506536724507, + "grad_norm": 1.166696349195129, + "learning_rate": 9.481369228080611e-07, + "loss": 0.5458, + "step": 3394 + }, + { + "epoch": 0.8069883527454242, + "grad_norm": 1.1763789566553926, + "learning_rate": 9.458823572608494e-07, + "loss": 0.5112, + "step": 3395 + }, + { + "epoch": 0.8072260518183979, + "grad_norm": 1.1394792398405167, + "learning_rate": 9.436301953886268e-07, + "loss": 0.5402, + "step": 3396 + }, + { + "epoch": 0.8074637508913716, + "grad_norm": 1.161760177347166, + "learning_rate": 9.41380438526694e-07, + "loss": 0.5499, + "step": 3397 + }, + { + "epoch": 0.8077014499643451, + "grad_norm": 1.3759434694243133, + "learning_rate": 9.391330880089261e-07, + "loss": 0.5816, + "step": 3398 + }, + { + "epoch": 0.8079391490373188, + "grad_norm": 1.3005627324592288, + "learning_rate": 9.368881451677725e-07, + "loss": 0.58, + "step": 3399 + }, + { + "epoch": 0.8081768481102923, + "grad_norm": 1.466299160039138, + "learning_rate": 9.346456113342506e-07, + "loss": 0.5812, + "step": 3400 + }, + { + "epoch": 0.8081768481102923, + "eval_loss": 0.5565537214279175, + "eval_runtime": 279.4458, + "eval_samples_per_second": 38.927, + "eval_steps_per_second": 1.217, + "step": 3400 + }, + { + "epoch": 0.808414547183266, + "grad_norm": 1.125766499517254, + "learning_rate": 9.324054878379535e-07, + "loss": 0.5608, + "step": 3401 + }, + { + "epoch": 0.8086522462562395, + "grad_norm": 1.147549794601611, + "learning_rate": 9.301677760070449e-07, + "loss": 0.5791, + "step": 3402 + }, + { + "epoch": 0.8088899453292132, + "grad_norm": 1.2914793744353348, + "learning_rate": 9.279324771682586e-07, + "loss": 0.5532, + "step": 3403 + }, + { + "epoch": 0.8091276444021869, + "grad_norm": 1.1900986232820876, + "learning_rate": 9.256995926468942e-07, + "loss": 0.5477, + "step": 3404 + }, + { + "epoch": 0.8093653434751604, + "grad_norm": 1.4616138636989822, + "learning_rate": 9.234691237668247e-07, + "loss": 0.4921, + "step": 3405 + }, + { + "epoch": 0.8096030425481341, + "grad_norm": 1.2832727945113056, + "learning_rate": 9.212410718504883e-07, + "loss": 0.5268, + "step": 3406 + }, + { + "epoch": 0.8098407416211076, + "grad_norm": 1.1640133250457971, + "learning_rate": 9.190154382188921e-07, + "loss": 0.5505, + "step": 3407 + }, + { + "epoch": 0.8100784406940813, + "grad_norm": 1.28950337324173, + "learning_rate": 9.167922241916055e-07, + "loss": 0.5383, + "step": 3408 + }, + { + "epoch": 0.8103161397670549, + "grad_norm": 1.2482695984465098, + "learning_rate": 9.145714310867676e-07, + "loss": 0.5642, + "step": 3409 + }, + { + "epoch": 0.8105538388400285, + "grad_norm": 1.1574372117230727, + "learning_rate": 9.123530602210795e-07, + "loss": 0.5691, + "step": 3410 + }, + { + "epoch": 0.8107915379130022, + "grad_norm": 1.268947984796699, + "learning_rate": 9.101371129098091e-07, + "loss": 0.5531, + "step": 3411 + }, + { + "epoch": 0.8110292369859757, + "grad_norm": 1.3545750030111348, + "learning_rate": 9.079235904667826e-07, + "loss": 0.5521, + "step": 3412 + }, + { + "epoch": 0.8112669360589494, + "grad_norm": 1.2873686487752565, + "learning_rate": 9.057124942043926e-07, + "loss": 0.5542, + "step": 3413 + }, + { + "epoch": 0.811504635131923, + "grad_norm": 1.3709959859675234, + "learning_rate": 9.03503825433592e-07, + "loss": 0.5385, + "step": 3414 + }, + { + "epoch": 0.8117423342048966, + "grad_norm": 1.3077523084122202, + "learning_rate": 9.01297585463895e-07, + "loss": 0.5758, + "step": 3415 + }, + { + "epoch": 0.8119800332778702, + "grad_norm": 1.6046652103848191, + "learning_rate": 8.99093775603373e-07, + "loss": 0.5311, + "step": 3416 + }, + { + "epoch": 0.8122177323508438, + "grad_norm": 1.211290900980416, + "learning_rate": 8.968923971586596e-07, + "loss": 0.5471, + "step": 3417 + }, + { + "epoch": 0.8124554314238175, + "grad_norm": 1.3391338635003351, + "learning_rate": 8.946934514349465e-07, + "loss": 0.5486, + "step": 3418 + }, + { + "epoch": 0.812693130496791, + "grad_norm": 1.2947285853095667, + "learning_rate": 8.924969397359834e-07, + "loss": 0.515, + "step": 3419 + }, + { + "epoch": 0.8129308295697647, + "grad_norm": 1.7026781456879734, + "learning_rate": 8.903028633640737e-07, + "loss": 0.5899, + "step": 3420 + }, + { + "epoch": 0.8131685286427383, + "grad_norm": 1.270735973141611, + "learning_rate": 8.881112236200795e-07, + "loss": 0.5619, + "step": 3421 + }, + { + "epoch": 0.8134062277157119, + "grad_norm": 1.4135235129305375, + "learning_rate": 8.859220218034192e-07, + "loss": 0.533, + "step": 3422 + }, + { + "epoch": 0.8136439267886856, + "grad_norm": 1.2428676003052623, + "learning_rate": 8.837352592120646e-07, + "loss": 0.5458, + "step": 3423 + }, + { + "epoch": 0.8138816258616591, + "grad_norm": 1.0978454880379764, + "learning_rate": 8.815509371425385e-07, + "loss": 0.5952, + "step": 3424 + }, + { + "epoch": 0.8141193249346328, + "grad_norm": 1.2533513813095918, + "learning_rate": 8.793690568899216e-07, + "loss": 0.4744, + "step": 3425 + }, + { + "epoch": 0.8143570240076063, + "grad_norm": 1.2749011570895379, + "learning_rate": 8.771896197478441e-07, + "loss": 0.586, + "step": 3426 + }, + { + "epoch": 0.81459472308058, + "grad_norm": 1.297734807926823, + "learning_rate": 8.750126270084891e-07, + "loss": 0.5606, + "step": 3427 + }, + { + "epoch": 0.8148324221535536, + "grad_norm": 1.3071384582157, + "learning_rate": 8.728380799625874e-07, + "loss": 0.5801, + "step": 3428 + }, + { + "epoch": 0.8150701212265272, + "grad_norm": 1.3721037291708524, + "learning_rate": 8.706659798994227e-07, + "loss": 0.4962, + "step": 3429 + }, + { + "epoch": 0.8153078202995009, + "grad_norm": 1.4168791491393762, + "learning_rate": 8.684963281068276e-07, + "loss": 0.5555, + "step": 3430 + }, + { + "epoch": 0.8155455193724744, + "grad_norm": 1.2395314672970728, + "learning_rate": 8.663291258711831e-07, + "loss": 0.5653, + "step": 3431 + }, + { + "epoch": 0.8157832184454481, + "grad_norm": 1.2388788881807402, + "learning_rate": 8.641643744774159e-07, + "loss": 0.5517, + "step": 3432 + }, + { + "epoch": 0.8160209175184217, + "grad_norm": 1.2934197135477419, + "learning_rate": 8.620020752090008e-07, + "loss": 0.5201, + "step": 3433 + }, + { + "epoch": 0.8162586165913953, + "grad_norm": 1.3981126544206495, + "learning_rate": 8.598422293479603e-07, + "loss": 0.546, + "step": 3434 + }, + { + "epoch": 0.8164963156643689, + "grad_norm": 1.311844177669402, + "learning_rate": 8.57684838174862e-07, + "loss": 0.5504, + "step": 3435 + }, + { + "epoch": 0.8167340147373425, + "grad_norm": 1.1317569011008515, + "learning_rate": 8.555299029688141e-07, + "loss": 0.5522, + "step": 3436 + }, + { + "epoch": 0.8169717138103162, + "grad_norm": 1.3096827579866572, + "learning_rate": 8.533774250074727e-07, + "loss": 0.5588, + "step": 3437 + }, + { + "epoch": 0.8172094128832897, + "grad_norm": 1.2727709497047819, + "learning_rate": 8.512274055670367e-07, + "loss": 0.571, + "step": 3438 + }, + { + "epoch": 0.8174471119562634, + "grad_norm": 1.1373375162726023, + "learning_rate": 8.490798459222477e-07, + "loss": 0.5224, + "step": 3439 + }, + { + "epoch": 0.817684811029237, + "grad_norm": 1.3870993056898715, + "learning_rate": 8.469347473463846e-07, + "loss": 0.5951, + "step": 3440 + }, + { + "epoch": 0.8179225101022106, + "grad_norm": 1.234831459799233, + "learning_rate": 8.447921111112722e-07, + "loss": 0.5436, + "step": 3441 + }, + { + "epoch": 0.8181602091751842, + "grad_norm": 1.1821592479057002, + "learning_rate": 8.426519384872733e-07, + "loss": 0.5439, + "step": 3442 + }, + { + "epoch": 0.8183979082481578, + "grad_norm": 1.3615221470899954, + "learning_rate": 8.405142307432917e-07, + "loss": 0.5463, + "step": 3443 + }, + { + "epoch": 0.8186356073211315, + "grad_norm": 1.3232071099815348, + "learning_rate": 8.383789891467664e-07, + "loss": 0.5651, + "step": 3444 + }, + { + "epoch": 0.818873306394105, + "grad_norm": 1.163879912051638, + "learning_rate": 8.362462149636757e-07, + "loss": 0.5266, + "step": 3445 + }, + { + "epoch": 0.8191110054670787, + "grad_norm": 1.049669327212281, + "learning_rate": 8.341159094585383e-07, + "loss": 0.4884, + "step": 3446 + }, + { + "epoch": 0.8193487045400523, + "grad_norm": 1.235056426238154, + "learning_rate": 8.31988073894403e-07, + "loss": 0.557, + "step": 3447 + }, + { + "epoch": 0.8195864036130259, + "grad_norm": 1.8957187333543728, + "learning_rate": 8.298627095328593e-07, + "loss": 0.5466, + "step": 3448 + }, + { + "epoch": 0.8198241026859995, + "grad_norm": 1.2026353704300308, + "learning_rate": 8.277398176340295e-07, + "loss": 0.5722, + "step": 3449 + }, + { + "epoch": 0.8200618017589731, + "grad_norm": 1.9904910163999605, + "learning_rate": 8.25619399456572e-07, + "loss": 0.5874, + "step": 3450 + }, + { + "epoch": 0.8202995008319468, + "grad_norm": 1.358905135247961, + "learning_rate": 8.235014562576732e-07, + "loss": 0.5673, + "step": 3451 + }, + { + "epoch": 0.8205371999049204, + "grad_norm": 1.2600263028404373, + "learning_rate": 8.213859892930581e-07, + "loss": 0.5375, + "step": 3452 + }, + { + "epoch": 0.820774898977894, + "grad_norm": 1.2680160140199568, + "learning_rate": 8.192729998169801e-07, + "loss": 0.5739, + "step": 3453 + }, + { + "epoch": 0.8210125980508676, + "grad_norm": 1.43848319628474, + "learning_rate": 8.171624890822266e-07, + "loss": 0.5952, + "step": 3454 + }, + { + "epoch": 0.8212502971238412, + "grad_norm": 1.2054616677173646, + "learning_rate": 8.150544583401116e-07, + "loss": 0.5648, + "step": 3455 + }, + { + "epoch": 0.8214879961968148, + "grad_norm": 1.2539045327573666, + "learning_rate": 8.129489088404808e-07, + "loss": 0.5105, + "step": 3456 + }, + { + "epoch": 0.8217256952697884, + "grad_norm": 1.2555149610080742, + "learning_rate": 8.108458418317089e-07, + "loss": 0.6072, + "step": 3457 + }, + { + "epoch": 0.8219633943427621, + "grad_norm": 1.2332802683044561, + "learning_rate": 8.087452585606997e-07, + "loss": 0.5877, + "step": 3458 + }, + { + "epoch": 0.8222010934157357, + "grad_norm": 1.132768980796218, + "learning_rate": 8.066471602728804e-07, + "loss": 0.5345, + "step": 3459 + }, + { + "epoch": 0.8224387924887093, + "grad_norm": 1.2103317884158589, + "learning_rate": 8.045515482122096e-07, + "loss": 0.568, + "step": 3460 + }, + { + "epoch": 0.8226764915616829, + "grad_norm": 1.1998524373271626, + "learning_rate": 8.024584236211697e-07, + "loss": 0.5414, + "step": 3461 + }, + { + "epoch": 0.8229141906346565, + "grad_norm": 1.2848695405489596, + "learning_rate": 8.00367787740769e-07, + "loss": 0.5429, + "step": 3462 + }, + { + "epoch": 0.8231518897076301, + "grad_norm": 1.2122731331126562, + "learning_rate": 7.98279641810537e-07, + "loss": 0.518, + "step": 3463 + }, + { + "epoch": 0.8233895887806038, + "grad_norm": 1.6360082097859467, + "learning_rate": 7.961939870685321e-07, + "loss": 0.5937, + "step": 3464 + }, + { + "epoch": 0.8236272878535774, + "grad_norm": 1.250582210252881, + "learning_rate": 7.941108247513318e-07, + "loss": 0.5795, + "step": 3465 + }, + { + "epoch": 0.823864986926551, + "grad_norm": 1.7926302613944967, + "learning_rate": 7.920301560940391e-07, + "loss": 0.5208, + "step": 3466 + }, + { + "epoch": 0.8241026859995246, + "grad_norm": 1.1747357395637923, + "learning_rate": 7.899519823302743e-07, + "loss": 0.5363, + "step": 3467 + }, + { + "epoch": 0.8243403850724982, + "grad_norm": 1.3031319083112758, + "learning_rate": 7.878763046921817e-07, + "loss": 0.5466, + "step": 3468 + }, + { + "epoch": 0.8245780841454718, + "grad_norm": 1.2392860272920783, + "learning_rate": 7.858031244104247e-07, + "loss": 0.5487, + "step": 3469 + }, + { + "epoch": 0.8248157832184454, + "grad_norm": 1.2938722682454613, + "learning_rate": 7.837324427141873e-07, + "loss": 0.5814, + "step": 3470 + }, + { + "epoch": 0.8250534822914191, + "grad_norm": 1.2729478581965044, + "learning_rate": 7.816642608311692e-07, + "loss": 0.5136, + "step": 3471 + }, + { + "epoch": 0.8252911813643927, + "grad_norm": 1.2070966496153717, + "learning_rate": 7.7959857998759e-07, + "loss": 0.5433, + "step": 3472 + }, + { + "epoch": 0.8255288804373663, + "grad_norm": 1.340474050268808, + "learning_rate": 7.77535401408186e-07, + "loss": 0.5365, + "step": 3473 + }, + { + "epoch": 0.8257665795103399, + "grad_norm": 1.131917037355165, + "learning_rate": 7.754747263162115e-07, + "loss": 0.5548, + "step": 3474 + }, + { + "epoch": 0.8260042785833135, + "grad_norm": 1.4189854729025086, + "learning_rate": 7.734165559334327e-07, + "loss": 0.5478, + "step": 3475 + }, + { + "epoch": 0.8262419776562872, + "grad_norm": 1.1448455384810396, + "learning_rate": 7.71360891480134e-07, + "loss": 0.5276, + "step": 3476 + }, + { + "epoch": 0.8264796767292607, + "grad_norm": 1.2699039521279718, + "learning_rate": 7.693077341751138e-07, + "loss": 0.5866, + "step": 3477 + }, + { + "epoch": 0.8267173758022344, + "grad_norm": 1.2119124201647486, + "learning_rate": 7.672570852356837e-07, + "loss": 0.5853, + "step": 3478 + }, + { + "epoch": 0.826955074875208, + "grad_norm": 1.165348632451427, + "learning_rate": 7.652089458776651e-07, + "loss": 0.5481, + "step": 3479 + }, + { + "epoch": 0.8271927739481816, + "grad_norm": 1.3743802100895852, + "learning_rate": 7.631633173153963e-07, + "loss": 0.5174, + "step": 3480 + }, + { + "epoch": 0.8274304730211552, + "grad_norm": 1.2577378836636703, + "learning_rate": 7.611202007617241e-07, + "loss": 0.5015, + "step": 3481 + }, + { + "epoch": 0.8276681720941288, + "grad_norm": 1.148639825283776, + "learning_rate": 7.590795974280079e-07, + "loss": 0.5436, + "step": 3482 + }, + { + "epoch": 0.8279058711671025, + "grad_norm": 1.2265336463105903, + "learning_rate": 7.570415085241129e-07, + "loss": 0.5327, + "step": 3483 + }, + { + "epoch": 0.828143570240076, + "grad_norm": 1.2227483495128908, + "learning_rate": 7.550059352584182e-07, + "loss": 0.566, + "step": 3484 + }, + { + "epoch": 0.8283812693130497, + "grad_norm": 1.2329950793854074, + "learning_rate": 7.529728788378093e-07, + "loss": 0.5607, + "step": 3485 + }, + { + "epoch": 0.8286189683860233, + "grad_norm": 1.2523920648261175, + "learning_rate": 7.509423404676802e-07, + "loss": 0.5533, + "step": 3486 + }, + { + "epoch": 0.8288566674589969, + "grad_norm": 1.5281367993967685, + "learning_rate": 7.489143213519301e-07, + "loss": 0.6085, + "step": 3487 + }, + { + "epoch": 0.8290943665319706, + "grad_norm": 1.567871514900946, + "learning_rate": 7.46888822692966e-07, + "loss": 0.5481, + "step": 3488 + }, + { + "epoch": 0.8293320656049441, + "grad_norm": 1.2162278565887552, + "learning_rate": 7.448658456917013e-07, + "loss": 0.5435, + "step": 3489 + }, + { + "epoch": 0.8295697646779178, + "grad_norm": 1.1526964124441759, + "learning_rate": 7.428453915475542e-07, + "loss": 0.5641, + "step": 3490 + }, + { + "epoch": 0.8298074637508913, + "grad_norm": 1.284277376415709, + "learning_rate": 7.408274614584432e-07, + "loss": 0.5647, + "step": 3491 + }, + { + "epoch": 0.830045162823865, + "grad_norm": 1.1921747887767329, + "learning_rate": 7.388120566207957e-07, + "loss": 0.5513, + "step": 3492 + }, + { + "epoch": 0.8302828618968386, + "grad_norm": 1.095272336832619, + "learning_rate": 7.367991782295392e-07, + "loss": 0.5473, + "step": 3493 + }, + { + "epoch": 0.8305205609698122, + "grad_norm": 1.2845563924360301, + "learning_rate": 7.347888274781045e-07, + "loss": 0.5103, + "step": 3494 + }, + { + "epoch": 0.8307582600427859, + "grad_norm": 1.4784436382346333, + "learning_rate": 7.327810055584211e-07, + "loss": 0.5624, + "step": 3495 + }, + { + "epoch": 0.8309959591157594, + "grad_norm": 1.7525106778308717, + "learning_rate": 7.307757136609218e-07, + "loss": 0.5464, + "step": 3496 + }, + { + "epoch": 0.8312336581887331, + "grad_norm": 1.227662562296749, + "learning_rate": 7.287729529745386e-07, + "loss": 0.4857, + "step": 3497 + }, + { + "epoch": 0.8314713572617067, + "grad_norm": 1.3836195448569126, + "learning_rate": 7.26772724686704e-07, + "loss": 0.557, + "step": 3498 + }, + { + "epoch": 0.8317090563346803, + "grad_norm": 1.3009396451264954, + "learning_rate": 7.24775029983345e-07, + "loss": 0.593, + "step": 3499 + }, + { + "epoch": 0.831946755407654, + "grad_norm": 1.2649253335625625, + "learning_rate": 7.227798700488914e-07, + "loss": 0.5812, + "step": 3500 + }, + { + "epoch": 0.831946755407654, + "eval_loss": 0.5553056597709656, + "eval_runtime": 284.4165, + "eval_samples_per_second": 38.247, + "eval_steps_per_second": 1.195, + "step": 3500 + }, + { + "epoch": 0.8321844544806275, + "grad_norm": 1.1969234624887077, + "learning_rate": 7.207872460662663e-07, + "loss": 0.5146, + "step": 3501 + }, + { + "epoch": 0.8324221535536012, + "grad_norm": 1.4157289094129302, + "learning_rate": 7.187971592168936e-07, + "loss": 0.5525, + "step": 3502 + }, + { + "epoch": 0.8326598526265747, + "grad_norm": 1.3164673749662648, + "learning_rate": 7.168096106806871e-07, + "loss": 0.5567, + "step": 3503 + }, + { + "epoch": 0.8328975516995484, + "grad_norm": 1.2625579571187808, + "learning_rate": 7.148246016360599e-07, + "loss": 0.5335, + "step": 3504 + }, + { + "epoch": 0.833135250772522, + "grad_norm": 1.1215990211334148, + "learning_rate": 7.128421332599189e-07, + "loss": 0.5967, + "step": 3505 + }, + { + "epoch": 0.8333729498454956, + "grad_norm": 1.1175098057664798, + "learning_rate": 7.108622067276633e-07, + "loss": 0.536, + "step": 3506 + }, + { + "epoch": 0.8336106489184693, + "grad_norm": 1.351987424772332, + "learning_rate": 7.088848232131862e-07, + "loss": 0.5798, + "step": 3507 + }, + { + "epoch": 0.8338483479914428, + "grad_norm": 1.314548045258486, + "learning_rate": 7.069099838888733e-07, + "loss": 0.5537, + "step": 3508 + }, + { + "epoch": 0.8340860470644165, + "grad_norm": 1.2615704575636946, + "learning_rate": 7.049376899256017e-07, + "loss": 0.5744, + "step": 3509 + }, + { + "epoch": 0.83432374613739, + "grad_norm": 1.3550022329400244, + "learning_rate": 7.029679424927366e-07, + "loss": 0.5396, + "step": 3510 + }, + { + "epoch": 0.8345614452103637, + "grad_norm": 1.4806683096346929, + "learning_rate": 7.010007427581378e-07, + "loss": 0.539, + "step": 3511 + }, + { + "epoch": 0.8347991442833373, + "grad_norm": 1.2380977984554313, + "learning_rate": 6.990360918881517e-07, + "loss": 0.547, + "step": 3512 + }, + { + "epoch": 0.8350368433563109, + "grad_norm": 1.2887574059750488, + "learning_rate": 6.97073991047616e-07, + "loss": 0.6224, + "step": 3513 + }, + { + "epoch": 0.8352745424292846, + "grad_norm": 1.1084158705268663, + "learning_rate": 6.951144413998517e-07, + "loss": 0.5705, + "step": 3514 + }, + { + "epoch": 0.8355122415022581, + "grad_norm": 1.226573118043958, + "learning_rate": 6.931574441066725e-07, + "loss": 0.542, + "step": 3515 + }, + { + "epoch": 0.8357499405752318, + "grad_norm": 1.4453671640552166, + "learning_rate": 6.912030003283754e-07, + "loss": 0.5566, + "step": 3516 + }, + { + "epoch": 0.8359876396482053, + "grad_norm": 1.2404712785324474, + "learning_rate": 6.892511112237472e-07, + "loss": 0.5303, + "step": 3517 + }, + { + "epoch": 0.836225338721179, + "grad_norm": 1.271552332788637, + "learning_rate": 6.87301777950054e-07, + "loss": 0.544, + "step": 3518 + }, + { + "epoch": 0.8364630377941527, + "grad_norm": 1.460632970346306, + "learning_rate": 6.853550016630517e-07, + "loss": 0.5508, + "step": 3519 + }, + { + "epoch": 0.8367007368671262, + "grad_norm": 1.2409969151293059, + "learning_rate": 6.834107835169784e-07, + "loss": 0.5626, + "step": 3520 + }, + { + "epoch": 0.8369384359400999, + "grad_norm": 1.523939834296756, + "learning_rate": 6.814691246645566e-07, + "loss": 0.5736, + "step": 3521 + }, + { + "epoch": 0.8371761350130734, + "grad_norm": 1.1765739946534395, + "learning_rate": 6.795300262569882e-07, + "loss": 0.5461, + "step": 3522 + }, + { + "epoch": 0.8374138340860471, + "grad_norm": 1.2571089677405887, + "learning_rate": 6.775934894439606e-07, + "loss": 0.5743, + "step": 3523 + }, + { + "epoch": 0.8376515331590206, + "grad_norm": 1.0813809435309676, + "learning_rate": 6.756595153736412e-07, + "loss": 0.5349, + "step": 3524 + }, + { + "epoch": 0.8378892322319943, + "grad_norm": 1.2514860734723368, + "learning_rate": 6.737281051926781e-07, + "loss": 0.5597, + "step": 3525 + }, + { + "epoch": 0.838126931304968, + "grad_norm": 1.523433078620397, + "learning_rate": 6.71799260046197e-07, + "loss": 0.5554, + "step": 3526 + }, + { + "epoch": 0.8383646303779415, + "grad_norm": 1.4363434694516268, + "learning_rate": 6.698729810778065e-07, + "loss": 0.5554, + "step": 3527 + }, + { + "epoch": 0.8386023294509152, + "grad_norm": 1.1433884136264543, + "learning_rate": 6.679492694295919e-07, + "loss": 0.5574, + "step": 3528 + }, + { + "epoch": 0.8388400285238887, + "grad_norm": 1.671228034396523, + "learning_rate": 6.66028126242117e-07, + "loss": 0.5484, + "step": 3529 + }, + { + "epoch": 0.8390777275968624, + "grad_norm": 2.4751426021426, + "learning_rate": 6.6410955265442e-07, + "loss": 0.5566, + "step": 3530 + }, + { + "epoch": 0.8393154266698359, + "grad_norm": 1.3622611007786307, + "learning_rate": 6.6219354980402e-07, + "loss": 0.5245, + "step": 3531 + }, + { + "epoch": 0.8395531257428096, + "grad_norm": 1.0135760494653607, + "learning_rate": 6.602801188269081e-07, + "loss": 0.5071, + "step": 3532 + }, + { + "epoch": 0.8397908248157833, + "grad_norm": 1.0646367151861056, + "learning_rate": 6.583692608575543e-07, + "loss": 0.5161, + "step": 3533 + }, + { + "epoch": 0.8400285238887568, + "grad_norm": 1.723990939405067, + "learning_rate": 6.564609770288987e-07, + "loss": 0.5855, + "step": 3534 + }, + { + "epoch": 0.8402662229617305, + "grad_norm": 1.1671417969069708, + "learning_rate": 6.545552684723583e-07, + "loss": 0.5149, + "step": 3535 + }, + { + "epoch": 0.840503922034704, + "grad_norm": 1.4108095456837948, + "learning_rate": 6.526521363178224e-07, + "loss": 0.5449, + "step": 3536 + }, + { + "epoch": 0.8407416211076777, + "grad_norm": 1.742113054923295, + "learning_rate": 6.507515816936538e-07, + "loss": 0.5425, + "step": 3537 + }, + { + "epoch": 0.8409793201806512, + "grad_norm": 1.3412680490652606, + "learning_rate": 6.488536057266842e-07, + "loss": 0.5741, + "step": 3538 + }, + { + "epoch": 0.8412170192536249, + "grad_norm": 1.227386304274121, + "learning_rate": 6.469582095422183e-07, + "loss": 0.547, + "step": 3539 + }, + { + "epoch": 0.8414547183265986, + "grad_norm": 1.1944138282542345, + "learning_rate": 6.450653942640323e-07, + "loss": 0.5787, + "step": 3540 + }, + { + "epoch": 0.8416924173995721, + "grad_norm": 1.0878812587190678, + "learning_rate": 6.431751610143716e-07, + "loss": 0.5074, + "step": 3541 + }, + { + "epoch": 0.8419301164725458, + "grad_norm": 1.6806408667929287, + "learning_rate": 6.412875109139482e-07, + "loss": 0.526, + "step": 3542 + }, + { + "epoch": 0.8421678155455193, + "grad_norm": 1.076592980814473, + "learning_rate": 6.394024450819458e-07, + "loss": 0.5594, + "step": 3543 + }, + { + "epoch": 0.842405514618493, + "grad_norm": 1.3747009188606474, + "learning_rate": 6.375199646360142e-07, + "loss": 0.5802, + "step": 3544 + }, + { + "epoch": 0.8426432136914666, + "grad_norm": 1.3263917087907746, + "learning_rate": 6.356400706922722e-07, + "loss": 0.5425, + "step": 3545 + }, + { + "epoch": 0.8428809127644402, + "grad_norm": 1.3822261598300822, + "learning_rate": 6.337627643653011e-07, + "loss": 0.5497, + "step": 3546 + }, + { + "epoch": 0.8431186118374139, + "grad_norm": 1.4989572805720217, + "learning_rate": 6.318880467681527e-07, + "loss": 0.5305, + "step": 3547 + }, + { + "epoch": 0.8433563109103874, + "grad_norm": 1.2666380833928563, + "learning_rate": 6.3001591901234e-07, + "loss": 0.5439, + "step": 3548 + }, + { + "epoch": 0.8435940099833611, + "grad_norm": 1.1611967573458153, + "learning_rate": 6.281463822078454e-07, + "loss": 0.5017, + "step": 3549 + }, + { + "epoch": 0.8438317090563346, + "grad_norm": 1.1767583559859793, + "learning_rate": 6.262794374631082e-07, + "loss": 0.5003, + "step": 3550 + }, + { + "epoch": 0.8440694081293083, + "grad_norm": 1.349760736397175, + "learning_rate": 6.244150858850368e-07, + "loss": 0.5263, + "step": 3551 + }, + { + "epoch": 0.8443071072022819, + "grad_norm": 1.3123247832465421, + "learning_rate": 6.225533285789997e-07, + "loss": 0.5556, + "step": 3552 + }, + { + "epoch": 0.8445448062752555, + "grad_norm": 1.3826271689159966, + "learning_rate": 6.206941666488287e-07, + "loss": 0.6327, + "step": 3553 + }, + { + "epoch": 0.8447825053482292, + "grad_norm": 1.1387890044668243, + "learning_rate": 6.188376011968133e-07, + "loss": 0.5876, + "step": 3554 + }, + { + "epoch": 0.8450202044212027, + "grad_norm": 1.253208721792236, + "learning_rate": 6.169836333237073e-07, + "loss": 0.5187, + "step": 3555 + }, + { + "epoch": 0.8452579034941764, + "grad_norm": 1.3403659445230947, + "learning_rate": 6.151322641287233e-07, + "loss": 0.5542, + "step": 3556 + }, + { + "epoch": 0.84549560256715, + "grad_norm": 1.303163948609892, + "learning_rate": 6.132834947095334e-07, + "loss": 0.5793, + "step": 3557 + }, + { + "epoch": 0.8457333016401236, + "grad_norm": 1.2062304068200107, + "learning_rate": 6.114373261622659e-07, + "loss": 0.524, + "step": 3558 + }, + { + "epoch": 0.8459710007130972, + "grad_norm": 1.2039052456303845, + "learning_rate": 6.095937595815104e-07, + "loss": 0.5825, + "step": 3559 + }, + { + "epoch": 0.8462086997860708, + "grad_norm": 1.2684360886774886, + "learning_rate": 6.077527960603119e-07, + "loss": 0.5537, + "step": 3560 + }, + { + "epoch": 0.8464463988590445, + "grad_norm": 1.3577519463820875, + "learning_rate": 6.059144366901737e-07, + "loss": 0.5403, + "step": 3561 + }, + { + "epoch": 0.846684097932018, + "grad_norm": 1.2922705912311199, + "learning_rate": 6.040786825610518e-07, + "loss": 0.5129, + "step": 3562 + }, + { + "epoch": 0.8469217970049917, + "grad_norm": 1.5745610916839212, + "learning_rate": 6.022455347613609e-07, + "loss": 0.4892, + "step": 3563 + }, + { + "epoch": 0.8471594960779653, + "grad_norm": 1.4183936564259323, + "learning_rate": 6.004149943779692e-07, + "loss": 0.5492, + "step": 3564 + }, + { + "epoch": 0.8473971951509389, + "grad_norm": 1.2122765055308111, + "learning_rate": 5.985870624961993e-07, + "loss": 0.5572, + "step": 3565 + }, + { + "epoch": 0.8476348942239125, + "grad_norm": 1.3637688436622768, + "learning_rate": 5.96761740199826e-07, + "loss": 0.538, + "step": 3566 + }, + { + "epoch": 0.8478725932968861, + "grad_norm": 1.2949738147634795, + "learning_rate": 5.949390285710777e-07, + "loss": 0.5157, + "step": 3567 + }, + { + "epoch": 0.8481102923698598, + "grad_norm": 1.2907790332234585, + "learning_rate": 5.931189286906358e-07, + "loss": 0.5351, + "step": 3568 + }, + { + "epoch": 0.8483479914428333, + "grad_norm": 1.236615513431733, + "learning_rate": 5.913014416376328e-07, + "loss": 0.5579, + "step": 3569 + }, + { + "epoch": 0.848585690515807, + "grad_norm": 1.294733140131259, + "learning_rate": 5.894865684896501e-07, + "loss": 0.5705, + "step": 3570 + }, + { + "epoch": 0.8488233895887806, + "grad_norm": 1.3085639621781227, + "learning_rate": 5.876743103227217e-07, + "loss": 0.5054, + "step": 3571 + }, + { + "epoch": 0.8490610886617542, + "grad_norm": 1.3194740632541317, + "learning_rate": 5.858646682113306e-07, + "loss": 0.5531, + "step": 3572 + }, + { + "epoch": 0.8492987877347279, + "grad_norm": 1.2183639495732546, + "learning_rate": 5.840576432284096e-07, + "loss": 0.5517, + "step": 3573 + }, + { + "epoch": 0.8495364868077014, + "grad_norm": 1.334887165891189, + "learning_rate": 5.822532364453365e-07, + "loss": 0.5516, + "step": 3574 + }, + { + "epoch": 0.8497741858806751, + "grad_norm": 1.3500880211145405, + "learning_rate": 5.804514489319402e-07, + "loss": 0.6032, + "step": 3575 + }, + { + "epoch": 0.8500118849536487, + "grad_norm": 1.2107051235202084, + "learning_rate": 5.786522817564955e-07, + "loss": 0.5342, + "step": 3576 + }, + { + "epoch": 0.8502495840266223, + "grad_norm": 1.3908906521288886, + "learning_rate": 5.768557359857241e-07, + "loss": 0.6024, + "step": 3577 + }, + { + "epoch": 0.8504872830995959, + "grad_norm": 1.2138871429586486, + "learning_rate": 5.750618126847912e-07, + "loss": 0.5194, + "step": 3578 + }, + { + "epoch": 0.8507249821725695, + "grad_norm": 1.306668121417796, + "learning_rate": 5.7327051291731e-07, + "loss": 0.5394, + "step": 3579 + }, + { + "epoch": 0.8509626812455432, + "grad_norm": 1.2574391317607296, + "learning_rate": 5.714818377453374e-07, + "loss": 0.5724, + "step": 3580 + }, + { + "epoch": 0.8512003803185167, + "grad_norm": 1.4405376862600672, + "learning_rate": 5.696957882293746e-07, + "loss": 0.5655, + "step": 3581 + }, + { + "epoch": 0.8514380793914904, + "grad_norm": 1.2205520901125657, + "learning_rate": 5.67912365428363e-07, + "loss": 0.5566, + "step": 3582 + }, + { + "epoch": 0.851675778464464, + "grad_norm": 1.1789050886777968, + "learning_rate": 5.661315703996905e-07, + "loss": 0.5147, + "step": 3583 + }, + { + "epoch": 0.8519134775374376, + "grad_norm": 1.3143333796046959, + "learning_rate": 5.643534041991855e-07, + "loss": 0.5235, + "step": 3584 + }, + { + "epoch": 0.8521511766104112, + "grad_norm": 1.252404666591287, + "learning_rate": 5.625778678811184e-07, + "loss": 0.4928, + "step": 3585 + }, + { + "epoch": 0.8523888756833848, + "grad_norm": 1.1445051001020157, + "learning_rate": 5.608049624981982e-07, + "loss": 0.5501, + "step": 3586 + }, + { + "epoch": 0.8526265747563585, + "grad_norm": 1.155144573047995, + "learning_rate": 5.590346891015758e-07, + "loss": 0.5345, + "step": 3587 + }, + { + "epoch": 0.852864273829332, + "grad_norm": 1.2474836084480396, + "learning_rate": 5.572670487408422e-07, + "loss": 0.5646, + "step": 3588 + }, + { + "epoch": 0.8531019729023057, + "grad_norm": 2.0134545558812635, + "learning_rate": 5.555020424640267e-07, + "loss": 0.5829, + "step": 3589 + }, + { + "epoch": 0.8533396719752793, + "grad_norm": 1.2774783533739893, + "learning_rate": 5.537396713175947e-07, + "loss": 0.5634, + "step": 3590 + }, + { + "epoch": 0.8535773710482529, + "grad_norm": 1.4779766560373842, + "learning_rate": 5.519799363464523e-07, + "loss": 0.5543, + "step": 3591 + }, + { + "epoch": 0.8538150701212265, + "grad_norm": 1.2328176809432645, + "learning_rate": 5.502228385939418e-07, + "loss": 0.5392, + "step": 3592 + }, + { + "epoch": 0.8540527691942001, + "grad_norm": 1.3768454224605196, + "learning_rate": 5.484683791018419e-07, + "loss": 0.6145, + "step": 3593 + }, + { + "epoch": 0.8542904682671738, + "grad_norm": 1.1821841622364058, + "learning_rate": 5.467165589103646e-07, + "loss": 0.5138, + "step": 3594 + }, + { + "epoch": 0.8545281673401474, + "grad_norm": 1.282370814217491, + "learning_rate": 5.449673790581611e-07, + "loss": 0.5445, + "step": 3595 + }, + { + "epoch": 0.854765866413121, + "grad_norm": 1.4790017098443562, + "learning_rate": 5.432208405823141e-07, + "loss": 0.531, + "step": 3596 + }, + { + "epoch": 0.8550035654860946, + "grad_norm": 1.258809654056054, + "learning_rate": 5.414769445183432e-07, + "loss": 0.5542, + "step": 3597 + }, + { + "epoch": 0.8552412645590682, + "grad_norm": 1.1766991854678375, + "learning_rate": 5.397356919001973e-07, + "loss": 0.5721, + "step": 3598 + }, + { + "epoch": 0.8554789636320418, + "grad_norm": 1.2178433625664504, + "learning_rate": 5.379970837602611e-07, + "loss": 0.537, + "step": 3599 + }, + { + "epoch": 0.8557166627050155, + "grad_norm": 1.3832815564121352, + "learning_rate": 5.362611211293506e-07, + "loss": 0.5598, + "step": 3600 + }, + { + "epoch": 0.8557166627050155, + "eval_loss": 0.5541449785232544, + "eval_runtime": 284.4403, + "eval_samples_per_second": 38.244, + "eval_steps_per_second": 1.195, + "step": 3600 + }, + { + "epoch": 0.8559543617779891, + "grad_norm": 1.390135353389405, + "learning_rate": 5.345278050367142e-07, + "loss": 0.5558, + "step": 3601 + }, + { + "epoch": 0.8561920608509627, + "grad_norm": 1.0979842372107416, + "learning_rate": 5.327971365100276e-07, + "loss": 0.5412, + "step": 3602 + }, + { + "epoch": 0.8564297599239363, + "grad_norm": 1.2776567210904681, + "learning_rate": 5.310691165754012e-07, + "loss": 0.5403, + "step": 3603 + }, + { + "epoch": 0.8566674589969099, + "grad_norm": 1.3695379349287378, + "learning_rate": 5.293437462573725e-07, + "loss": 0.5116, + "step": 3604 + }, + { + "epoch": 0.8569051580698835, + "grad_norm": 1.2200393303316295, + "learning_rate": 5.276210265789094e-07, + "loss": 0.5831, + "step": 3605 + }, + { + "epoch": 0.8571428571428571, + "grad_norm": 1.2183082399569745, + "learning_rate": 5.259009585614072e-07, + "loss": 0.5929, + "step": 3606 + }, + { + "epoch": 0.8573805562158308, + "grad_norm": 1.3746693837777688, + "learning_rate": 5.241835432246888e-07, + "loss": 0.5541, + "step": 3607 + }, + { + "epoch": 0.8576182552888044, + "grad_norm": 1.1538702063395576, + "learning_rate": 5.224687815870055e-07, + "loss": 0.5609, + "step": 3608 + }, + { + "epoch": 0.857855954361778, + "grad_norm": 1.3702312138631447, + "learning_rate": 5.207566746650355e-07, + "loss": 0.5499, + "step": 3609 + }, + { + "epoch": 0.8580936534347516, + "grad_norm": 1.309892017254969, + "learning_rate": 5.190472234738809e-07, + "loss": 0.4722, + "step": 3610 + }, + { + "epoch": 0.8583313525077252, + "grad_norm": 1.9707469058454747, + "learning_rate": 5.173404290270706e-07, + "loss": 0.5742, + "step": 3611 + }, + { + "epoch": 0.8585690515806988, + "grad_norm": 1.2112447030924542, + "learning_rate": 5.156362923365587e-07, + "loss": 0.5592, + "step": 3612 + }, + { + "epoch": 0.8588067506536724, + "grad_norm": 1.1787506003228987, + "learning_rate": 5.139348144127237e-07, + "loss": 0.5804, + "step": 3613 + }, + { + "epoch": 0.8590444497266461, + "grad_norm": 1.3918463257962297, + "learning_rate": 5.122359962643658e-07, + "loss": 0.5111, + "step": 3614 + }, + { + "epoch": 0.8592821487996197, + "grad_norm": 1.228980174614741, + "learning_rate": 5.105398388987098e-07, + "loss": 0.5702, + "step": 3615 + }, + { + "epoch": 0.8595198478725933, + "grad_norm": 1.4529110967764003, + "learning_rate": 5.08846343321403e-07, + "loss": 0.5202, + "step": 3616 + }, + { + "epoch": 0.8597575469455669, + "grad_norm": 1.2306540818406324, + "learning_rate": 5.071555105365156e-07, + "loss": 0.5578, + "step": 3617 + }, + { + "epoch": 0.8599952460185405, + "grad_norm": 1.2586737442356282, + "learning_rate": 5.054673415465344e-07, + "loss": 0.5556, + "step": 3618 + }, + { + "epoch": 0.8602329450915142, + "grad_norm": 1.218276236259285, + "learning_rate": 5.037818373523723e-07, + "loss": 0.5283, + "step": 3619 + }, + { + "epoch": 0.8604706441644877, + "grad_norm": 1.18480376129879, + "learning_rate": 5.020989989533592e-07, + "loss": 0.4934, + "step": 3620 + }, + { + "epoch": 0.8607083432374614, + "grad_norm": 1.1403709390561658, + "learning_rate": 5.004188273472471e-07, + "loss": 0.5411, + "step": 3621 + }, + { + "epoch": 0.860946042310435, + "grad_norm": 1.252929428862168, + "learning_rate": 4.987413235302025e-07, + "loss": 0.5593, + "step": 3622 + }, + { + "epoch": 0.8611837413834086, + "grad_norm": 1.271316270975205, + "learning_rate": 4.970664884968135e-07, + "loss": 0.5009, + "step": 3623 + }, + { + "epoch": 0.8614214404563822, + "grad_norm": 1.5309235292925836, + "learning_rate": 4.953943232400848e-07, + "loss": 0.5511, + "step": 3624 + }, + { + "epoch": 0.8616591395293558, + "grad_norm": 1.10723546065252, + "learning_rate": 4.937248287514407e-07, + "loss": 0.5412, + "step": 3625 + }, + { + "epoch": 0.8618968386023295, + "grad_norm": 1.3350081899342492, + "learning_rate": 4.92058006020717e-07, + "loss": 0.5505, + "step": 3626 + }, + { + "epoch": 0.862134537675303, + "grad_norm": 1.2958396065236049, + "learning_rate": 4.903938560361698e-07, + "loss": 0.5011, + "step": 3627 + }, + { + "epoch": 0.8623722367482767, + "grad_norm": 1.0940716658498821, + "learning_rate": 4.887323797844678e-07, + "loss": 0.5594, + "step": 3628 + }, + { + "epoch": 0.8626099358212503, + "grad_norm": 1.4767835564498848, + "learning_rate": 4.87073578250698e-07, + "loss": 0.5602, + "step": 3629 + }, + { + "epoch": 0.8628476348942239, + "grad_norm": 1.871459758871421, + "learning_rate": 4.854174524183564e-07, + "loss": 0.5709, + "step": 3630 + }, + { + "epoch": 0.8630853339671976, + "grad_norm": 1.376545235071449, + "learning_rate": 4.837640032693558e-07, + "loss": 0.5356, + "step": 3631 + }, + { + "epoch": 0.8633230330401711, + "grad_norm": 1.2482201638109496, + "learning_rate": 4.821132317840233e-07, + "loss": 0.5009, + "step": 3632 + }, + { + "epoch": 0.8635607321131448, + "grad_norm": 1.6502835586941187, + "learning_rate": 4.804651389410958e-07, + "loss": 0.6303, + "step": 3633 + }, + { + "epoch": 0.8637984311861183, + "grad_norm": 1.2964332335785376, + "learning_rate": 4.788197257177212e-07, + "loss": 0.5243, + "step": 3634 + }, + { + "epoch": 0.864036130259092, + "grad_norm": 1.3280711308869386, + "learning_rate": 4.771769930894632e-07, + "loss": 0.5403, + "step": 3635 + }, + { + "epoch": 0.8642738293320656, + "grad_norm": 1.285794219220523, + "learning_rate": 4.7553694203029245e-07, + "loss": 0.5413, + "step": 3636 + }, + { + "epoch": 0.8645115284050392, + "grad_norm": 1.6233379901331775, + "learning_rate": 4.738995735125895e-07, + "loss": 0.5841, + "step": 3637 + }, + { + "epoch": 0.8647492274780129, + "grad_norm": 1.2720909675212348, + "learning_rate": 4.722648885071468e-07, + "loss": 0.5135, + "step": 3638 + }, + { + "epoch": 0.8649869265509864, + "grad_norm": 1.182998767435298, + "learning_rate": 4.7063288798316397e-07, + "loss": 0.5572, + "step": 3639 + }, + { + "epoch": 0.8652246256239601, + "grad_norm": 1.3012709039442276, + "learning_rate": 4.6900357290825037e-07, + "loss": 0.4817, + "step": 3640 + }, + { + "epoch": 0.8654623246969336, + "grad_norm": 1.2925802021610935, + "learning_rate": 4.673769442484211e-07, + "loss": 0.5255, + "step": 3641 + }, + { + "epoch": 0.8657000237699073, + "grad_norm": 1.309344703623191, + "learning_rate": 4.6575300296809956e-07, + "loss": 0.5209, + "step": 3642 + }, + { + "epoch": 0.865937722842881, + "grad_norm": 1.4520859407476394, + "learning_rate": 4.641317500301173e-07, + "loss": 0.5701, + "step": 3643 + }, + { + "epoch": 0.8661754219158545, + "grad_norm": 1.0507118903644679, + "learning_rate": 4.625131863957111e-07, + "loss": 0.5521, + "step": 3644 + }, + { + "epoch": 0.8664131209888282, + "grad_norm": 1.206266028307072, + "learning_rate": 4.6089731302452033e-07, + "loss": 0.5225, + "step": 3645 + }, + { + "epoch": 0.8666508200618017, + "grad_norm": 1.2410447283390647, + "learning_rate": 4.5928413087459325e-07, + "loss": 0.5674, + "step": 3646 + }, + { + "epoch": 0.8668885191347754, + "grad_norm": 1.3885024057580742, + "learning_rate": 4.576736409023813e-07, + "loss": 0.5638, + "step": 3647 + }, + { + "epoch": 0.867126218207749, + "grad_norm": 1.27271956312215, + "learning_rate": 4.5606584406273967e-07, + "loss": 0.5768, + "step": 3648 + }, + { + "epoch": 0.8673639172807226, + "grad_norm": 1.509694916486589, + "learning_rate": 4.5446074130892525e-07, + "loss": 0.4617, + "step": 3649 + }, + { + "epoch": 0.8676016163536963, + "grad_norm": 1.3655743913834668, + "learning_rate": 4.528583335926001e-07, + "loss": 0.5852, + "step": 3650 + }, + { + "epoch": 0.8678393154266698, + "grad_norm": 2.1214216283520546, + "learning_rate": 4.512586218638265e-07, + "loss": 0.5914, + "step": 3651 + }, + { + "epoch": 0.8680770144996435, + "grad_norm": 1.3302636815531041, + "learning_rate": 4.4966160707107075e-07, + "loss": 0.5347, + "step": 3652 + }, + { + "epoch": 0.868314713572617, + "grad_norm": 1.2528477908984113, + "learning_rate": 4.4806729016119654e-07, + "loss": 0.5195, + "step": 3653 + }, + { + "epoch": 0.8685524126455907, + "grad_norm": 1.1953072640955396, + "learning_rate": 4.4647567207946994e-07, + "loss": 0.5455, + "step": 3654 + }, + { + "epoch": 0.8687901117185644, + "grad_norm": 1.2166206286444665, + "learning_rate": 4.448867537695578e-07, + "loss": 0.5635, + "step": 3655 + }, + { + "epoch": 0.8690278107915379, + "grad_norm": 1.2440669983357728, + "learning_rate": 4.4330053617352597e-07, + "loss": 0.5014, + "step": 3656 + }, + { + "epoch": 0.8692655098645116, + "grad_norm": 1.2365052880566523, + "learning_rate": 4.4171702023183663e-07, + "loss": 0.5285, + "step": 3657 + }, + { + "epoch": 0.8695032089374851, + "grad_norm": 1.3129563836619242, + "learning_rate": 4.401362068833526e-07, + "loss": 0.5158, + "step": 3658 + }, + { + "epoch": 0.8697409080104588, + "grad_norm": 1.106400986977446, + "learning_rate": 4.3855809706533357e-07, + "loss": 0.5363, + "step": 3659 + }, + { + "epoch": 0.8699786070834323, + "grad_norm": 1.1780577932839367, + "learning_rate": 4.3698269171343844e-07, + "loss": 0.5359, + "step": 3660 + }, + { + "epoch": 0.870216306156406, + "grad_norm": 1.1341751082454816, + "learning_rate": 4.3540999176171717e-07, + "loss": 0.5382, + "step": 3661 + }, + { + "epoch": 0.8704540052293797, + "grad_norm": 1.279572311160812, + "learning_rate": 4.338399981426211e-07, + "loss": 0.563, + "step": 3662 + }, + { + "epoch": 0.8706917043023532, + "grad_norm": 1.2670785535878042, + "learning_rate": 4.322727117869951e-07, + "loss": 0.5654, + "step": 3663 + }, + { + "epoch": 0.8709294033753269, + "grad_norm": 1.2141056760454505, + "learning_rate": 4.3070813362407924e-07, + "loss": 0.5712, + "step": 3664 + }, + { + "epoch": 0.8711671024483004, + "grad_norm": 1.4801984319435648, + "learning_rate": 4.2914626458150644e-07, + "loss": 0.5419, + "step": 3665 + }, + { + "epoch": 0.8714048015212741, + "grad_norm": 1.1387806446617643, + "learning_rate": 4.275871055853042e-07, + "loss": 0.5076, + "step": 3666 + }, + { + "epoch": 0.8716425005942476, + "grad_norm": 1.6428589730032723, + "learning_rate": 4.2603065755989493e-07, + "loss": 0.5689, + "step": 3667 + }, + { + "epoch": 0.8718801996672213, + "grad_norm": 1.502607713400422, + "learning_rate": 4.2447692142809193e-07, + "loss": 0.5762, + "step": 3668 + }, + { + "epoch": 0.872117898740195, + "grad_norm": 1.28049838669455, + "learning_rate": 4.2292589811110006e-07, + "loss": 0.5737, + "step": 3669 + }, + { + "epoch": 0.8723555978131685, + "grad_norm": 1.3851006617354427, + "learning_rate": 4.213775885285176e-07, + "loss": 0.5336, + "step": 3670 + }, + { + "epoch": 0.8725932968861422, + "grad_norm": 1.0887787883831512, + "learning_rate": 4.198319935983325e-07, + "loss": 0.5293, + "step": 3671 + }, + { + "epoch": 0.8728309959591157, + "grad_norm": 1.4443847792520368, + "learning_rate": 4.182891142369244e-07, + "loss": 0.5813, + "step": 3672 + }, + { + "epoch": 0.8730686950320894, + "grad_norm": 1.4290034413860904, + "learning_rate": 4.167489513590611e-07, + "loss": 0.4749, + "step": 3673 + }, + { + "epoch": 0.8733063941050629, + "grad_norm": 1.2112427437650424, + "learning_rate": 4.1521150587790103e-07, + "loss": 0.5278, + "step": 3674 + }, + { + "epoch": 0.8735440931780366, + "grad_norm": 1.250364327075875, + "learning_rate": 4.1367677870499166e-07, + "loss": 0.5561, + "step": 3675 + }, + { + "epoch": 0.8737817922510103, + "grad_norm": 1.1878840988352155, + "learning_rate": 4.1214477075026957e-07, + "loss": 0.546, + "step": 3676 + }, + { + "epoch": 0.8740194913239838, + "grad_norm": 1.444888901282434, + "learning_rate": 4.10615482922056e-07, + "loss": 0.5113, + "step": 3677 + }, + { + "epoch": 0.8742571903969575, + "grad_norm": 1.1916530900409612, + "learning_rate": 4.090889161270617e-07, + "loss": 0.5447, + "step": 3678 + }, + { + "epoch": 0.874494889469931, + "grad_norm": 1.2781343072106517, + "learning_rate": 4.0756507127038494e-07, + "loss": 0.5595, + "step": 3679 + }, + { + "epoch": 0.8747325885429047, + "grad_norm": 1.1561959637261328, + "learning_rate": 4.0604394925550906e-07, + "loss": 0.5571, + "step": 3680 + }, + { + "epoch": 0.8749702876158783, + "grad_norm": 1.1238897648668986, + "learning_rate": 4.0452555098430156e-07, + "loss": 0.5493, + "step": 3681 + }, + { + "epoch": 0.8752079866888519, + "grad_norm": 1.2099842434434613, + "learning_rate": 4.0300987735701733e-07, + "loss": 0.5703, + "step": 3682 + }, + { + "epoch": 0.8754456857618256, + "grad_norm": 1.2845655092332737, + "learning_rate": 4.0149692927229533e-07, + "loss": 0.5428, + "step": 3683 + }, + { + "epoch": 0.8756833848347991, + "grad_norm": 1.249321919739779, + "learning_rate": 3.9998670762715907e-07, + "loss": 0.5171, + "step": 3684 + }, + { + "epoch": 0.8759210839077728, + "grad_norm": 1.4636289149187074, + "learning_rate": 3.984792133170129e-07, + "loss": 0.5319, + "step": 3685 + }, + { + "epoch": 0.8761587829807463, + "grad_norm": 1.2742534169094961, + "learning_rate": 3.969744472356463e-07, + "loss": 0.5316, + "step": 3686 + }, + { + "epoch": 0.87639648205372, + "grad_norm": 1.386077903502547, + "learning_rate": 3.9547241027523164e-07, + "loss": 0.5664, + "step": 3687 + }, + { + "epoch": 0.8766341811266936, + "grad_norm": 1.351903249661024, + "learning_rate": 3.939731033263228e-07, + "loss": 0.5413, + "step": 3688 + }, + { + "epoch": 0.8768718801996672, + "grad_norm": 1.494534464705481, + "learning_rate": 3.924765272778536e-07, + "loss": 0.5229, + "step": 3689 + }, + { + "epoch": 0.8771095792726409, + "grad_norm": 1.2512742752001338, + "learning_rate": 3.9098268301713983e-07, + "loss": 0.5894, + "step": 3690 + }, + { + "epoch": 0.8773472783456144, + "grad_norm": 1.494123267993252, + "learning_rate": 3.894915714298775e-07, + "loss": 0.5182, + "step": 3691 + }, + { + "epoch": 0.8775849774185881, + "grad_norm": 1.3509874397235806, + "learning_rate": 3.8800319340014437e-07, + "loss": 0.5794, + "step": 3692 + }, + { + "epoch": 0.8778226764915616, + "grad_norm": 1.3721011034603061, + "learning_rate": 3.865175498103929e-07, + "loss": 0.5596, + "step": 3693 + }, + { + "epoch": 0.8780603755645353, + "grad_norm": 1.285312993821372, + "learning_rate": 3.850346415414585e-07, + "loss": 0.545, + "step": 3694 + }, + { + "epoch": 0.8782980746375089, + "grad_norm": 1.2446753974704083, + "learning_rate": 3.8355446947255293e-07, + "loss": 0.5246, + "step": 3695 + }, + { + "epoch": 0.8785357737104825, + "grad_norm": 1.2027094691887026, + "learning_rate": 3.820770344812674e-07, + "loss": 0.4967, + "step": 3696 + }, + { + "epoch": 0.8787734727834562, + "grad_norm": 1.1672112788393156, + "learning_rate": 3.8060233744356634e-07, + "loss": 0.5379, + "step": 3697 + }, + { + "epoch": 0.8790111718564297, + "grad_norm": 1.3314367660463962, + "learning_rate": 3.7913037923379526e-07, + "loss": 0.5856, + "step": 3698 + }, + { + "epoch": 0.8792488709294034, + "grad_norm": 1.4394394685096077, + "learning_rate": 3.776611607246744e-07, + "loss": 0.5491, + "step": 3699 + }, + { + "epoch": 0.879486570002377, + "grad_norm": 1.2962696143329258, + "learning_rate": 3.761946827872981e-07, + "loss": 0.5726, + "step": 3700 + }, + { + "epoch": 0.879486570002377, + "eval_loss": 0.5533133149147034, + "eval_runtime": 284.6782, + "eval_samples_per_second": 38.212, + "eval_steps_per_second": 1.194, + "step": 3700 + }, + { + "epoch": 0.8797242690753506, + "grad_norm": 1.2032785516539748, + "learning_rate": 3.7473094629113746e-07, + "loss": 0.5378, + "step": 3701 + }, + { + "epoch": 0.8799619681483242, + "grad_norm": 1.211897481165327, + "learning_rate": 3.732699521040378e-07, + "loss": 0.5208, + "step": 3702 + }, + { + "epoch": 0.8801996672212978, + "grad_norm": 1.2717889018143964, + "learning_rate": 3.71811701092219e-07, + "loss": 0.549, + "step": 3703 + }, + { + "epoch": 0.8804373662942715, + "grad_norm": 1.6020604982510098, + "learning_rate": 3.703561941202732e-07, + "loss": 0.5788, + "step": 3704 + }, + { + "epoch": 0.880675065367245, + "grad_norm": 1.1904059937228428, + "learning_rate": 3.689034320511664e-07, + "loss": 0.5038, + "step": 3705 + }, + { + "epoch": 0.8809127644402187, + "grad_norm": 1.4234122848884305, + "learning_rate": 3.6745341574623785e-07, + "loss": 0.6056, + "step": 3706 + }, + { + "epoch": 0.8811504635131923, + "grad_norm": 1.1656093639907719, + "learning_rate": 3.660061460651981e-07, + "loss": 0.5442, + "step": 3707 + }, + { + "epoch": 0.8813881625861659, + "grad_norm": 1.4483420076212472, + "learning_rate": 3.6456162386612813e-07, + "loss": 0.5607, + "step": 3708 + }, + { + "epoch": 0.8816258616591395, + "grad_norm": 1.166665396859563, + "learning_rate": 3.6311985000548223e-07, + "loss": 0.5554, + "step": 3709 + }, + { + "epoch": 0.8818635607321131, + "grad_norm": 1.3487297181617286, + "learning_rate": 3.616808253380838e-07, + "loss": 0.5323, + "step": 3710 + }, + { + "epoch": 0.8821012598050868, + "grad_norm": 1.5863063032075242, + "learning_rate": 3.602445507171276e-07, + "loss": 0.4953, + "step": 3711 + }, + { + "epoch": 0.8823389588780604, + "grad_norm": 1.2926198267026419, + "learning_rate": 3.588110269941747e-07, + "loss": 0.5466, + "step": 3712 + }, + { + "epoch": 0.882576657951034, + "grad_norm": 1.2376904011008856, + "learning_rate": 3.5738025501915794e-07, + "loss": 0.564, + "step": 3713 + }, + { + "epoch": 0.8828143570240076, + "grad_norm": 1.2833962209797702, + "learning_rate": 3.5595223564037884e-07, + "loss": 0.5228, + "step": 3714 + }, + { + "epoch": 0.8830520560969812, + "grad_norm": 2.011744679207707, + "learning_rate": 3.5452696970450674e-07, + "loss": 0.5312, + "step": 3715 + }, + { + "epoch": 0.8832897551699548, + "grad_norm": 1.7488315557971206, + "learning_rate": 3.531044580565751e-07, + "loss": 0.562, + "step": 3716 + }, + { + "epoch": 0.8835274542429284, + "grad_norm": 1.6962069944701723, + "learning_rate": 3.5168470153998937e-07, + "loss": 0.5571, + "step": 3717 + }, + { + "epoch": 0.8837651533159021, + "grad_norm": 1.3087506886584572, + "learning_rate": 3.5026770099651885e-07, + "loss": 0.5478, + "step": 3718 + }, + { + "epoch": 0.8840028523888757, + "grad_norm": 1.1887074993146598, + "learning_rate": 3.488534572662994e-07, + "loss": 0.5387, + "step": 3719 + }, + { + "epoch": 0.8842405514618493, + "grad_norm": 1.1924275327377039, + "learning_rate": 3.474419711878307e-07, + "loss": 0.5096, + "step": 3720 + }, + { + "epoch": 0.8844782505348229, + "grad_norm": 1.1503840034825183, + "learning_rate": 3.4603324359798016e-07, + "loss": 0.5482, + "step": 3721 + }, + { + "epoch": 0.8847159496077965, + "grad_norm": 1.243001335246571, + "learning_rate": 3.4462727533197794e-07, + "loss": 0.5466, + "step": 3722 + }, + { + "epoch": 0.8849536486807702, + "grad_norm": 1.3570163234139834, + "learning_rate": 3.4322406722341993e-07, + "loss": 0.5248, + "step": 3723 + }, + { + "epoch": 0.8851913477537438, + "grad_norm": 1.2777148346990403, + "learning_rate": 3.4182362010426184e-07, + "loss": 0.544, + "step": 3724 + }, + { + "epoch": 0.8854290468267174, + "grad_norm": 1.3574085555940594, + "learning_rate": 3.404259348048267e-07, + "loss": 0.5236, + "step": 3725 + }, + { + "epoch": 0.885666745899691, + "grad_norm": 1.6019569639041795, + "learning_rate": 3.390310121537982e-07, + "loss": 0.5779, + "step": 3726 + }, + { + "epoch": 0.8859044449726646, + "grad_norm": 1.1302417864133425, + "learning_rate": 3.3763885297822153e-07, + "loss": 0.6006, + "step": 3727 + }, + { + "epoch": 0.8861421440456382, + "grad_norm": 1.398067850718201, + "learning_rate": 3.362494581035036e-07, + "loss": 0.5547, + "step": 3728 + }, + { + "epoch": 0.8863798431186118, + "grad_norm": 1.5997214126464643, + "learning_rate": 3.3486282835341334e-07, + "loss": 0.52, + "step": 3729 + }, + { + "epoch": 0.8866175421915855, + "grad_norm": 1.3078113214582296, + "learning_rate": 3.3347896455007965e-07, + "loss": 0.5596, + "step": 3730 + }, + { + "epoch": 0.8868552412645591, + "grad_norm": 1.4294357207154378, + "learning_rate": 3.320978675139919e-07, + "loss": 0.5338, + "step": 3731 + }, + { + "epoch": 0.8870929403375327, + "grad_norm": 1.2314260438655458, + "learning_rate": 3.307195380639977e-07, + "loss": 0.5259, + "step": 3732 + }, + { + "epoch": 0.8873306394105063, + "grad_norm": 1.1242316887978687, + "learning_rate": 3.293439770173046e-07, + "loss": 0.5155, + "step": 3733 + }, + { + "epoch": 0.8875683384834799, + "grad_norm": 2.1990053030611767, + "learning_rate": 3.279711851894796e-07, + "loss": 0.5101, + "step": 3734 + }, + { + "epoch": 0.8878060375564535, + "grad_norm": 1.269671194760233, + "learning_rate": 3.266011633944477e-07, + "loss": 0.5135, + "step": 3735 + }, + { + "epoch": 0.8880437366294271, + "grad_norm": 1.1906991282592523, + "learning_rate": 3.2523391244448923e-07, + "loss": 0.5482, + "step": 3736 + }, + { + "epoch": 0.8882814357024008, + "grad_norm": 1.247825938742005, + "learning_rate": 3.238694331502451e-07, + "loss": 0.5521, + "step": 3737 + }, + { + "epoch": 0.8885191347753744, + "grad_norm": 1.4685960466093118, + "learning_rate": 3.225077263207099e-07, + "loss": 0.4907, + "step": 3738 + }, + { + "epoch": 0.888756833848348, + "grad_norm": 1.2878650692422156, + "learning_rate": 3.2114879276323783e-07, + "loss": 0.5745, + "step": 3739 + }, + { + "epoch": 0.8889945329213216, + "grad_norm": 1.232034536317365, + "learning_rate": 3.197926332835344e-07, + "loss": 0.5613, + "step": 3740 + }, + { + "epoch": 0.8892322319942952, + "grad_norm": 1.236709155103951, + "learning_rate": 3.1843924868566355e-07, + "loss": 0.5274, + "step": 3741 + }, + { + "epoch": 0.8894699310672688, + "grad_norm": 1.2761527671731778, + "learning_rate": 3.170886397720435e-07, + "loss": 0.571, + "step": 3742 + }, + { + "epoch": 0.8897076301402425, + "grad_norm": 1.2449725816881232, + "learning_rate": 3.1574080734344757e-07, + "loss": 0.5519, + "step": 3743 + }, + { + "epoch": 0.8899453292132161, + "grad_norm": 1.3250465697266505, + "learning_rate": 3.1439575219899986e-07, + "loss": 0.5648, + "step": 3744 + }, + { + "epoch": 0.8901830282861897, + "grad_norm": 1.177537054952248, + "learning_rate": 3.130534751361808e-07, + "loss": 0.5088, + "step": 3745 + }, + { + "epoch": 0.8904207273591633, + "grad_norm": 1.19214232506579, + "learning_rate": 3.117139769508226e-07, + "loss": 0.5727, + "step": 3746 + }, + { + "epoch": 0.8906584264321369, + "grad_norm": 1.386557275834653, + "learning_rate": 3.103772584371106e-07, + "loss": 0.5093, + "step": 3747 + }, + { + "epoch": 0.8908961255051105, + "grad_norm": 1.4307790574862036, + "learning_rate": 3.0904332038757977e-07, + "loss": 0.5398, + "step": 3748 + }, + { + "epoch": 0.8911338245780841, + "grad_norm": 1.1866962095090776, + "learning_rate": 3.0771216359311963e-07, + "loss": 0.494, + "step": 3749 + }, + { + "epoch": 0.8913715236510578, + "grad_norm": 1.472164735721535, + "learning_rate": 3.0638378884296883e-07, + "loss": 0.5405, + "step": 3750 + }, + { + "epoch": 0.8916092227240314, + "grad_norm": 1.1802966237521972, + "learning_rate": 3.0505819692471797e-07, + "loss": 0.5599, + "step": 3751 + }, + { + "epoch": 0.891846921797005, + "grad_norm": 1.202385268519403, + "learning_rate": 3.037353886243055e-07, + "loss": 0.5691, + "step": 3752 + }, + { + "epoch": 0.8920846208699786, + "grad_norm": 1.387259448291775, + "learning_rate": 3.0241536472602084e-07, + "loss": 0.5367, + "step": 3753 + }, + { + "epoch": 0.8923223199429522, + "grad_norm": 1.4501085179610147, + "learning_rate": 3.010981260125029e-07, + "loss": 0.6045, + "step": 3754 + }, + { + "epoch": 0.8925600190159259, + "grad_norm": 1.2724229816186916, + "learning_rate": 2.9978367326473966e-07, + "loss": 0.5519, + "step": 3755 + }, + { + "epoch": 0.8927977180888994, + "grad_norm": 1.6683434428058872, + "learning_rate": 2.984720072620645e-07, + "loss": 0.5018, + "step": 3756 + }, + { + "epoch": 0.8930354171618731, + "grad_norm": 1.2538002995913593, + "learning_rate": 2.9716312878216194e-07, + "loss": 0.541, + "step": 3757 + }, + { + "epoch": 0.8932731162348467, + "grad_norm": 1.1655033190476067, + "learning_rate": 2.958570386010623e-07, + "loss": 0.5132, + "step": 3758 + }, + { + "epoch": 0.8935108153078203, + "grad_norm": 1.164568360716327, + "learning_rate": 2.9455373749314285e-07, + "loss": 0.5291, + "step": 3759 + }, + { + "epoch": 0.893748514380794, + "grad_norm": 1.161434488458057, + "learning_rate": 2.932532262311261e-07, + "loss": 0.6011, + "step": 3760 + }, + { + "epoch": 0.8939862134537675, + "grad_norm": 1.4182917755384186, + "learning_rate": 2.9195550558608176e-07, + "loss": 0.5139, + "step": 3761 + }, + { + "epoch": 0.8942239125267412, + "grad_norm": 1.2082900574208706, + "learning_rate": 2.9066057632742674e-07, + "loss": 0.5242, + "step": 3762 + }, + { + "epoch": 0.8944616115997147, + "grad_norm": 1.2836764710682154, + "learning_rate": 2.893684392229185e-07, + "loss": 0.5165, + "step": 3763 + }, + { + "epoch": 0.8946993106726884, + "grad_norm": 1.192476033546108, + "learning_rate": 2.8807909503866247e-07, + "loss": 0.5629, + "step": 3764 + }, + { + "epoch": 0.894937009745662, + "grad_norm": 1.197547462635026, + "learning_rate": 2.867925445391079e-07, + "loss": 0.5631, + "step": 3765 + }, + { + "epoch": 0.8951747088186356, + "grad_norm": 1.1612559004989853, + "learning_rate": 2.8550878848704666e-07, + "loss": 0.4993, + "step": 3766 + }, + { + "epoch": 0.8954124078916093, + "grad_norm": 1.3030994163534704, + "learning_rate": 2.842278276436128e-07, + "loss": 0.5597, + "step": 3767 + }, + { + "epoch": 0.8956501069645828, + "grad_norm": 1.6192622891280377, + "learning_rate": 2.8294966276828537e-07, + "loss": 0.5215, + "step": 3768 + }, + { + "epoch": 0.8958878060375565, + "grad_norm": 1.3638710080577325, + "learning_rate": 2.8167429461888496e-07, + "loss": 0.5262, + "step": 3769 + }, + { + "epoch": 0.89612550511053, + "grad_norm": 1.466618959699265, + "learning_rate": 2.804017239515744e-07, + "loss": 0.5639, + "step": 3770 + }, + { + "epoch": 0.8963632041835037, + "grad_norm": 1.3771399340478, + "learning_rate": 2.791319515208557e-07, + "loss": 0.5345, + "step": 3771 + }, + { + "epoch": 0.8966009032564773, + "grad_norm": 1.4762774555600986, + "learning_rate": 2.778649780795739e-07, + "loss": 0.5519, + "step": 3772 + }, + { + "epoch": 0.8968386023294509, + "grad_norm": 1.2504151471512701, + "learning_rate": 2.7660080437891477e-07, + "loss": 0.5316, + "step": 3773 + }, + { + "epoch": 0.8970763014024246, + "grad_norm": 1.2481553442538216, + "learning_rate": 2.753394311684032e-07, + "loss": 0.5184, + "step": 3774 + }, + { + "epoch": 0.8973140004753981, + "grad_norm": 1.3093859695110064, + "learning_rate": 2.7408085919590265e-07, + "loss": 0.5484, + "step": 3775 + }, + { + "epoch": 0.8975516995483718, + "grad_norm": 2.2328650354528086, + "learning_rate": 2.728250892076173e-07, + "loss": 0.5287, + "step": 3776 + }, + { + "epoch": 0.8977893986213453, + "grad_norm": 1.0982204132062963, + "learning_rate": 2.71572121948091e-07, + "loss": 0.5177, + "step": 3777 + }, + { + "epoch": 0.898027097694319, + "grad_norm": 1.6888397195191336, + "learning_rate": 2.703219581602035e-07, + "loss": 0.5475, + "step": 3778 + }, + { + "epoch": 0.8982647967672927, + "grad_norm": 1.276857350534146, + "learning_rate": 2.6907459858517336e-07, + "loss": 0.5337, + "step": 3779 + }, + { + "epoch": 0.8985024958402662, + "grad_norm": 1.2899214167795106, + "learning_rate": 2.678300439625564e-07, + "loss": 0.5787, + "step": 3780 + }, + { + "epoch": 0.8987401949132399, + "grad_norm": 1.2900938292156763, + "learning_rate": 2.6658829503024566e-07, + "loss": 0.5432, + "step": 3781 + }, + { + "epoch": 0.8989778939862134, + "grad_norm": 1.2234447613007757, + "learning_rate": 2.653493525244721e-07, + "loss": 0.5589, + "step": 3782 + }, + { + "epoch": 0.8992155930591871, + "grad_norm": 1.523368294163284, + "learning_rate": 2.6411321717979886e-07, + "loss": 0.5506, + "step": 3783 + }, + { + "epoch": 0.8994532921321606, + "grad_norm": 1.1660902072017372, + "learning_rate": 2.628798897291285e-07, + "loss": 0.5384, + "step": 3784 + }, + { + "epoch": 0.8996909912051343, + "grad_norm": 1.2323667852809925, + "learning_rate": 2.6164937090369713e-07, + "loss": 0.516, + "step": 3785 + }, + { + "epoch": 0.899928690278108, + "grad_norm": 1.3212306014197404, + "learning_rate": 2.6042166143307624e-07, + "loss": 0.5651, + "step": 3786 + }, + { + "epoch": 0.9001663893510815, + "grad_norm": 1.5261351755071781, + "learning_rate": 2.5919676204517073e-07, + "loss": 0.5965, + "step": 3787 + }, + { + "epoch": 0.9004040884240552, + "grad_norm": 1.2351826275443005, + "learning_rate": 2.579746734662203e-07, + "loss": 0.5598, + "step": 3788 + }, + { + "epoch": 0.9006417874970287, + "grad_norm": 1.5506140971648317, + "learning_rate": 2.5675539642079803e-07, + "loss": 0.557, + "step": 3789 + }, + { + "epoch": 0.9008794865700024, + "grad_norm": 1.2441933416252349, + "learning_rate": 2.555389316318102e-07, + "loss": 0.546, + "step": 3790 + }, + { + "epoch": 0.9011171856429759, + "grad_norm": 1.1835167076094129, + "learning_rate": 2.5432527982049424e-07, + "loss": 0.5454, + "step": 3791 + }, + { + "epoch": 0.9013548847159496, + "grad_norm": 1.3188641388613882, + "learning_rate": 2.531144417064213e-07, + "loss": 0.5635, + "step": 3792 + }, + { + "epoch": 0.9015925837889233, + "grad_norm": 1.3306988027874076, + "learning_rate": 2.5190641800749424e-07, + "loss": 0.5606, + "step": 3793 + }, + { + "epoch": 0.9018302828618968, + "grad_norm": 1.1379969294367018, + "learning_rate": 2.5070120943994746e-07, + "loss": 0.5604, + "step": 3794 + }, + { + "epoch": 0.9020679819348705, + "grad_norm": 1.3630940471665327, + "learning_rate": 2.4949881671834487e-07, + "loss": 0.564, + "step": 3795 + }, + { + "epoch": 0.902305681007844, + "grad_norm": 1.1983672327201, + "learning_rate": 2.4829924055558174e-07, + "loss": 0.5907, + "step": 3796 + }, + { + "epoch": 0.9025433800808177, + "grad_norm": 1.3903151847073851, + "learning_rate": 2.471024816628836e-07, + "loss": 0.5318, + "step": 3797 + }, + { + "epoch": 0.9027810791537914, + "grad_norm": 1.206996270000148, + "learning_rate": 2.459085407498063e-07, + "loss": 0.535, + "step": 3798 + }, + { + "epoch": 0.9030187782267649, + "grad_norm": 1.3925670953544191, + "learning_rate": 2.447174185242324e-07, + "loss": 0.5187, + "step": 3799 + }, + { + "epoch": 0.9032564772997386, + "grad_norm": 1.2566541683452637, + "learning_rate": 2.435291156923758e-07, + "loss": 0.4862, + "step": 3800 + }, + { + "epoch": 0.9032564772997386, + "eval_loss": 0.5525994896888733, + "eval_runtime": 284.8483, + "eval_samples_per_second": 38.189, + "eval_steps_per_second": 1.194, + "step": 3800 + }, + { + "epoch": 0.9034941763727121, + "grad_norm": 1.3142101696555966, + "learning_rate": 2.4234363295877806e-07, + "loss": 0.5565, + "step": 3801 + }, + { + "epoch": 0.9037318754456858, + "grad_norm": 1.222769108837061, + "learning_rate": 2.411609710263091e-07, + "loss": 0.52, + "step": 3802 + }, + { + "epoch": 0.9039695745186593, + "grad_norm": 1.114892744418346, + "learning_rate": 2.399811305961647e-07, + "loss": 0.5111, + "step": 3803 + }, + { + "epoch": 0.904207273591633, + "grad_norm": 1.3895219182411425, + "learning_rate": 2.3880411236786914e-07, + "loss": 0.5736, + "step": 3804 + }, + { + "epoch": 0.9044449726646067, + "grad_norm": 1.4715489888901911, + "learning_rate": 2.3762991703927375e-07, + "loss": 0.5824, + "step": 3805 + }, + { + "epoch": 0.9046826717375802, + "grad_norm": 1.2360885086426103, + "learning_rate": 2.36458545306556e-07, + "loss": 0.5476, + "step": 3806 + }, + { + "epoch": 0.9049203708105539, + "grad_norm": 1.1759130173618317, + "learning_rate": 2.3528999786421758e-07, + "loss": 0.5448, + "step": 3807 + }, + { + "epoch": 0.9051580698835274, + "grad_norm": 1.2175529608156435, + "learning_rate": 2.3412427540508763e-07, + "loss": 0.5208, + "step": 3808 + }, + { + "epoch": 0.9053957689565011, + "grad_norm": 1.2947184901341735, + "learning_rate": 2.3296137862031999e-07, + "loss": 0.5176, + "step": 3809 + }, + { + "epoch": 0.9056334680294746, + "grad_norm": 1.3067044497245264, + "learning_rate": 2.3180130819939362e-07, + "loss": 0.5301, + "step": 3810 + }, + { + "epoch": 0.9058711671024483, + "grad_norm": 1.3835861373707699, + "learning_rate": 2.3064406483010947e-07, + "loss": 0.4999, + "step": 3811 + }, + { + "epoch": 0.906108866175422, + "grad_norm": 1.5257435129991432, + "learning_rate": 2.2948964919859428e-07, + "loss": 0.5174, + "step": 3812 + }, + { + "epoch": 0.9063465652483955, + "grad_norm": 1.380322526753492, + "learning_rate": 2.283380619892983e-07, + "loss": 0.5409, + "step": 3813 + }, + { + "epoch": 0.9065842643213692, + "grad_norm": 1.151719880020792, + "learning_rate": 2.2718930388499537e-07, + "loss": 0.5175, + "step": 3814 + }, + { + "epoch": 0.9068219633943427, + "grad_norm": 1.2252219369560093, + "learning_rate": 2.2604337556677846e-07, + "loss": 0.5426, + "step": 3815 + }, + { + "epoch": 0.9070596624673164, + "grad_norm": 1.2364536674370366, + "learning_rate": 2.2490027771406686e-07, + "loss": 0.581, + "step": 3816 + }, + { + "epoch": 0.90729736154029, + "grad_norm": 1.1624581778371623, + "learning_rate": 2.237600110046001e-07, + "loss": 0.5112, + "step": 3817 + }, + { + "epoch": 0.9075350606132636, + "grad_norm": 1.514018298579978, + "learning_rate": 2.2262257611443904e-07, + "loss": 0.5096, + "step": 3818 + }, + { + "epoch": 0.9077727596862373, + "grad_norm": 1.3274777559559272, + "learning_rate": 2.2148797371796483e-07, + "loss": 0.5158, + "step": 3819 + }, + { + "epoch": 0.9080104587592108, + "grad_norm": 1.4265739810156113, + "learning_rate": 2.20356204487881e-07, + "loss": 0.5959, + "step": 3820 + }, + { + "epoch": 0.9082481578321845, + "grad_norm": 1.1149560950605273, + "learning_rate": 2.1922726909520963e-07, + "loss": 0.5721, + "step": 3821 + }, + { + "epoch": 0.908485856905158, + "grad_norm": 1.8448579674936225, + "learning_rate": 2.1810116820929427e-07, + "loss": 0.5561, + "step": 3822 + }, + { + "epoch": 0.9087235559781317, + "grad_norm": 1.6433374074729004, + "learning_rate": 2.1697790249779638e-07, + "loss": 0.5546, + "step": 3823 + }, + { + "epoch": 0.9089612550511053, + "grad_norm": 1.2532068135287528, + "learning_rate": 2.1585747262669653e-07, + "loss": 0.5597, + "step": 3824 + }, + { + "epoch": 0.9091989541240789, + "grad_norm": 1.9320835788901485, + "learning_rate": 2.147398792602956e-07, + "loss": 0.542, + "step": 3825 + }, + { + "epoch": 0.9094366531970526, + "grad_norm": 1.3150886195291196, + "learning_rate": 2.136251230612113e-07, + "loss": 0.4932, + "step": 3826 + }, + { + "epoch": 0.9096743522700261, + "grad_norm": 2.002691383994259, + "learning_rate": 2.1251320469037827e-07, + "loss": 0.5688, + "step": 3827 + }, + { + "epoch": 0.9099120513429998, + "grad_norm": 1.3839579210280208, + "learning_rate": 2.1140412480705085e-07, + "loss": 0.59, + "step": 3828 + }, + { + "epoch": 0.9101497504159733, + "grad_norm": 1.3434707480005648, + "learning_rate": 2.102978840687997e-07, + "loss": 0.4987, + "step": 3829 + }, + { + "epoch": 0.910387449488947, + "grad_norm": 1.3290489761124415, + "learning_rate": 2.0919448313151125e-07, + "loss": 0.5049, + "step": 3830 + }, + { + "epoch": 0.9106251485619206, + "grad_norm": 1.2345296204695704, + "learning_rate": 2.080939226493889e-07, + "loss": 0.5843, + "step": 3831 + }, + { + "epoch": 0.9108628476348942, + "grad_norm": 1.3942778180335706, + "learning_rate": 2.0699620327495174e-07, + "loss": 0.6024, + "step": 3832 + }, + { + "epoch": 0.9111005467078679, + "grad_norm": 1.1271351731559225, + "learning_rate": 2.0590132565903475e-07, + "loss": 0.5197, + "step": 3833 + }, + { + "epoch": 0.9113382457808414, + "grad_norm": 1.3254149065243634, + "learning_rate": 2.048092904507881e-07, + "loss": 0.547, + "step": 3834 + }, + { + "epoch": 0.9115759448538151, + "grad_norm": 1.2729981689812837, + "learning_rate": 2.0372009829767558e-07, + "loss": 0.5945, + "step": 3835 + }, + { + "epoch": 0.9118136439267887, + "grad_norm": 1.436832567190803, + "learning_rate": 2.0263374984547724e-07, + "loss": 0.5467, + "step": 3836 + }, + { + "epoch": 0.9120513429997623, + "grad_norm": 1.3261586319279026, + "learning_rate": 2.0155024573828452e-07, + "loss": 0.5944, + "step": 3837 + }, + { + "epoch": 0.9122890420727359, + "grad_norm": 1.2867332452024791, + "learning_rate": 2.004695866185058e-07, + "loss": 0.5503, + "step": 3838 + }, + { + "epoch": 0.9125267411457095, + "grad_norm": 1.3317139779444094, + "learning_rate": 1.9939177312685963e-07, + "loss": 0.5392, + "step": 3839 + }, + { + "epoch": 0.9127644402186832, + "grad_norm": 1.490889035156706, + "learning_rate": 1.983168059023788e-07, + "loss": 0.5591, + "step": 3840 + }, + { + "epoch": 0.9130021392916567, + "grad_norm": 1.2893892151698931, + "learning_rate": 1.9724468558240838e-07, + "loss": 0.5303, + "step": 3841 + }, + { + "epoch": 0.9132398383646304, + "grad_norm": 1.1568389248269189, + "learning_rate": 1.961754128026061e-07, + "loss": 0.5739, + "step": 3842 + }, + { + "epoch": 0.913477537437604, + "grad_norm": 1.2269227883900495, + "learning_rate": 1.951089881969398e-07, + "loss": 0.561, + "step": 3843 + }, + { + "epoch": 0.9137152365105776, + "grad_norm": 1.2032082543193996, + "learning_rate": 1.940454123976898e-07, + "loss": 0.5359, + "step": 3844 + }, + { + "epoch": 0.9139529355835512, + "grad_norm": 1.2778784752935226, + "learning_rate": 1.9298468603544786e-07, + "loss": 0.5656, + "step": 3845 + }, + { + "epoch": 0.9141906346565248, + "grad_norm": 1.2681573502248626, + "learning_rate": 1.9192680973911636e-07, + "loss": 0.5618, + "step": 3846 + }, + { + "epoch": 0.9144283337294985, + "grad_norm": 1.1953978999768695, + "learning_rate": 1.908717841359048e-07, + "loss": 0.5356, + "step": 3847 + }, + { + "epoch": 0.914666032802472, + "grad_norm": 1.325342972493933, + "learning_rate": 1.8981960985133662e-07, + "loss": 0.5454, + "step": 3848 + }, + { + "epoch": 0.9149037318754457, + "grad_norm": 1.1676799601976675, + "learning_rate": 1.8877028750924288e-07, + "loss": 0.5401, + "step": 3849 + }, + { + "epoch": 0.9151414309484193, + "grad_norm": 1.1469738019737505, + "learning_rate": 1.8772381773176417e-07, + "loss": 0.5522, + "step": 3850 + }, + { + "epoch": 0.9153791300213929, + "grad_norm": 1.2807876239088818, + "learning_rate": 1.8668020113934816e-07, + "loss": 0.5295, + "step": 3851 + }, + { + "epoch": 0.9156168290943665, + "grad_norm": 1.2410060099040048, + "learning_rate": 1.8563943835075315e-07, + "loss": 0.5224, + "step": 3852 + }, + { + "epoch": 0.9158545281673401, + "grad_norm": 1.43251718259575, + "learning_rate": 1.8460152998304393e-07, + "loss": 0.4933, + "step": 3853 + }, + { + "epoch": 0.9160922272403138, + "grad_norm": 1.2090898001354766, + "learning_rate": 1.835664766515949e-07, + "loss": 0.5272, + "step": 3854 + }, + { + "epoch": 0.9163299263132874, + "grad_norm": 1.3778221606100218, + "learning_rate": 1.825342789700846e-07, + "loss": 0.5496, + "step": 3855 + }, + { + "epoch": 0.916567625386261, + "grad_norm": 1.4029229149672917, + "learning_rate": 1.8150493755050068e-07, + "loss": 0.54, + "step": 3856 + }, + { + "epoch": 0.9168053244592346, + "grad_norm": 1.2906465353773386, + "learning_rate": 1.8047845300313726e-07, + "loss": 0.5294, + "step": 3857 + }, + { + "epoch": 0.9170430235322082, + "grad_norm": 1.1456769862614573, + "learning_rate": 1.7945482593659412e-07, + "loss": 0.5503, + "step": 3858 + }, + { + "epoch": 0.9172807226051818, + "grad_norm": 1.2147599346978863, + "learning_rate": 1.7843405695777582e-07, + "loss": 0.5439, + "step": 3859 + }, + { + "epoch": 0.9175184216781554, + "grad_norm": 1.230729919491286, + "learning_rate": 1.7741614667189433e-07, + "loss": 0.5539, + "step": 3860 + }, + { + "epoch": 0.9177561207511291, + "grad_norm": 1.3583079575611612, + "learning_rate": 1.7640109568246577e-07, + "loss": 0.5361, + "step": 3861 + }, + { + "epoch": 0.9179938198241027, + "grad_norm": 1.3659236338510403, + "learning_rate": 1.7538890459131098e-07, + "loss": 0.4952, + "step": 3862 + }, + { + "epoch": 0.9182315188970763, + "grad_norm": 1.1814149096202686, + "learning_rate": 1.7437957399855488e-07, + "loss": 0.555, + "step": 3863 + }, + { + "epoch": 0.9184692179700499, + "grad_norm": 1.1685137941433292, + "learning_rate": 1.733731045026271e-07, + "loss": 0.5264, + "step": 3864 + }, + { + "epoch": 0.9187069170430235, + "grad_norm": 1.4666780539547228, + "learning_rate": 1.7236949670026037e-07, + "loss": 0.5731, + "step": 3865 + }, + { + "epoch": 0.9189446161159971, + "grad_norm": 1.2456300330630006, + "learning_rate": 1.7136875118649143e-07, + "loss": 0.5101, + "step": 3866 + }, + { + "epoch": 0.9191823151889708, + "grad_norm": 1.154916995458485, + "learning_rate": 1.7037086855465902e-07, + "loss": 0.5451, + "step": 3867 + }, + { + "epoch": 0.9194200142619444, + "grad_norm": 1.4021465135718756, + "learning_rate": 1.6937584939640483e-07, + "loss": 0.5249, + "step": 3868 + }, + { + "epoch": 0.919657713334918, + "grad_norm": 1.2237529078812914, + "learning_rate": 1.6838369430167312e-07, + "loss": 0.5508, + "step": 3869 + }, + { + "epoch": 0.9198954124078916, + "grad_norm": 1.490943297578547, + "learning_rate": 1.6739440385871054e-07, + "loss": 0.6012, + "step": 3870 + }, + { + "epoch": 0.9201331114808652, + "grad_norm": 1.2093811559144703, + "learning_rate": 1.664079786540629e-07, + "loss": 0.544, + "step": 3871 + }, + { + "epoch": 0.9203708105538388, + "grad_norm": 1.1676086559228316, + "learning_rate": 1.6542441927258068e-07, + "loss": 0.5529, + "step": 3872 + }, + { + "epoch": 0.9206085096268125, + "grad_norm": 1.2266864491662368, + "learning_rate": 1.6444372629741246e-07, + "loss": 0.5767, + "step": 3873 + }, + { + "epoch": 0.9208462086997861, + "grad_norm": 1.4411627785394066, + "learning_rate": 1.6346590031000974e-07, + "loss": 0.5239, + "step": 3874 + }, + { + "epoch": 0.9210839077727597, + "grad_norm": 1.1992891349688783, + "learning_rate": 1.624909418901205e-07, + "loss": 0.5256, + "step": 3875 + }, + { + "epoch": 0.9213216068457333, + "grad_norm": 1.3266292488653866, + "learning_rate": 1.6151885161579673e-07, + "loss": 0.5786, + "step": 3876 + }, + { + "epoch": 0.9215593059187069, + "grad_norm": 1.2467001036345455, + "learning_rate": 1.6054963006338742e-07, + "loss": 0.5398, + "step": 3877 + }, + { + "epoch": 0.9217970049916805, + "grad_norm": 1.566785211677807, + "learning_rate": 1.595832778075418e-07, + "loss": 0.515, + "step": 3878 + }, + { + "epoch": 0.9220347040646542, + "grad_norm": 1.3195659775207493, + "learning_rate": 1.5861979542120598e-07, + "loss": 0.5182, + "step": 3879 + }, + { + "epoch": 0.9222724031376278, + "grad_norm": 1.3836450696148603, + "learning_rate": 1.5765918347562747e-07, + "loss": 0.5605, + "step": 3880 + }, + { + "epoch": 0.9225101022106014, + "grad_norm": 1.3382405586348853, + "learning_rate": 1.567014425403496e-07, + "loss": 0.5478, + "step": 3881 + }, + { + "epoch": 0.922747801283575, + "grad_norm": 1.327940495649445, + "learning_rate": 1.5574657318321528e-07, + "loss": 0.5452, + "step": 3882 + }, + { + "epoch": 0.9229855003565486, + "grad_norm": 1.2660925110971544, + "learning_rate": 1.547945759703623e-07, + "loss": 0.537, + "step": 3883 + }, + { + "epoch": 0.9232231994295222, + "grad_norm": 1.1939023597745397, + "learning_rate": 1.5384545146622854e-07, + "loss": 0.5373, + "step": 3884 + }, + { + "epoch": 0.9234608985024958, + "grad_norm": 1.2754668014207768, + "learning_rate": 1.528992002335461e-07, + "loss": 0.523, + "step": 3885 + }, + { + "epoch": 0.9236985975754695, + "grad_norm": 1.2623330181284051, + "learning_rate": 1.519558228333462e-07, + "loss": 0.566, + "step": 3886 + }, + { + "epoch": 0.9239362966484431, + "grad_norm": 1.1479437881954204, + "learning_rate": 1.510153198249531e-07, + "loss": 0.5597, + "step": 3887 + }, + { + "epoch": 0.9241739957214167, + "grad_norm": 1.6403053035810218, + "learning_rate": 1.5007769176598907e-07, + "loss": 0.5438, + "step": 3888 + }, + { + "epoch": 0.9244116947943903, + "grad_norm": 1.213946864455653, + "learning_rate": 1.491429392123711e-07, + "loss": 0.591, + "step": 3889 + }, + { + "epoch": 0.9246493938673639, + "grad_norm": 1.2232994069950154, + "learning_rate": 1.4821106271831143e-07, + "loss": 0.5551, + "step": 3890 + }, + { + "epoch": 0.9248870929403376, + "grad_norm": 1.2407447877867501, + "learning_rate": 1.4728206283631697e-07, + "loss": 0.5064, + "step": 3891 + }, + { + "epoch": 0.9251247920133111, + "grad_norm": 1.5695976502074045, + "learning_rate": 1.4635594011718935e-07, + "loss": 0.5577, + "step": 3892 + }, + { + "epoch": 0.9253624910862848, + "grad_norm": 1.224912900061704, + "learning_rate": 1.4543269511002435e-07, + "loss": 0.5129, + "step": 3893 + }, + { + "epoch": 0.9256001901592584, + "grad_norm": 1.1502816175651873, + "learning_rate": 1.4451232836221018e-07, + "loss": 0.5839, + "step": 3894 + }, + { + "epoch": 0.925837889232232, + "grad_norm": 1.1650207142529543, + "learning_rate": 1.435948404194304e-07, + "loss": 0.5426, + "step": 3895 + }, + { + "epoch": 0.9260755883052056, + "grad_norm": 1.2215303157604926, + "learning_rate": 1.4268023182566148e-07, + "loss": 0.5811, + "step": 3896 + }, + { + "epoch": 0.9263132873781792, + "grad_norm": 1.4130686068124174, + "learning_rate": 1.4176850312317246e-07, + "loss": 0.5476, + "step": 3897 + }, + { + "epoch": 0.9265509864511529, + "grad_norm": 1.190292754106079, + "learning_rate": 1.4085965485252373e-07, + "loss": 0.5418, + "step": 3898 + }, + { + "epoch": 0.9267886855241264, + "grad_norm": 1.3289387140476416, + "learning_rate": 1.399536875525692e-07, + "loss": 0.5671, + "step": 3899 + }, + { + "epoch": 0.9270263845971001, + "grad_norm": 1.190809495679312, + "learning_rate": 1.3905060176045427e-07, + "loss": 0.5076, + "step": 3900 + }, + { + "epoch": 0.9270263845971001, + "eval_loss": 0.5522435307502747, + "eval_runtime": 285.2928, + "eval_samples_per_second": 38.129, + "eval_steps_per_second": 1.192, + "step": 3900 + }, + { + "epoch": 0.9272640836700737, + "grad_norm": 1.2187533457694413, + "learning_rate": 1.3815039801161723e-07, + "loss": 0.4899, + "step": 3901 + }, + { + "epoch": 0.9275017827430473, + "grad_norm": 1.4659623060316052, + "learning_rate": 1.372530768397845e-07, + "loss": 0.5584, + "step": 3902 + }, + { + "epoch": 0.927739481816021, + "grad_norm": 1.2217054039259383, + "learning_rate": 1.363586387769761e-07, + "loss": 0.5216, + "step": 3903 + }, + { + "epoch": 0.9279771808889945, + "grad_norm": 1.391088250356047, + "learning_rate": 1.354670843535022e-07, + "loss": 0.5644, + "step": 3904 + }, + { + "epoch": 0.9282148799619682, + "grad_norm": 1.2438454581703566, + "learning_rate": 1.345784140979628e-07, + "loss": 0.5174, + "step": 3905 + }, + { + "epoch": 0.9284525790349417, + "grad_norm": 1.1847322735603585, + "learning_rate": 1.3369262853724696e-07, + "loss": 0.5131, + "step": 3906 + }, + { + "epoch": 0.9286902781079154, + "grad_norm": 1.5982763034931027, + "learning_rate": 1.328097281965357e-07, + "loss": 0.5788, + "step": 3907 + }, + { + "epoch": 0.928927977180889, + "grad_norm": 1.2266267331894376, + "learning_rate": 1.3192971359929752e-07, + "loss": 0.5241, + "step": 3908 + }, + { + "epoch": 0.9291656762538626, + "grad_norm": 1.339592898453164, + "learning_rate": 1.310525852672917e-07, + "loss": 0.55, + "step": 3909 + }, + { + "epoch": 0.9294033753268363, + "grad_norm": 1.2171988158365152, + "learning_rate": 1.3017834372056393e-07, + "loss": 0.5363, + "step": 3910 + }, + { + "epoch": 0.9296410743998098, + "grad_norm": 1.1233457447844848, + "learning_rate": 1.2930698947744957e-07, + "loss": 0.5472, + "step": 3911 + }, + { + "epoch": 0.9298787734727835, + "grad_norm": 1.187620691607735, + "learning_rate": 1.284385230545726e-07, + "loss": 0.6029, + "step": 3912 + }, + { + "epoch": 0.930116472545757, + "grad_norm": 1.3006265537731583, + "learning_rate": 1.2757294496684447e-07, + "loss": 0.5266, + "step": 3913 + }, + { + "epoch": 0.9303541716187307, + "grad_norm": 1.2074885430544644, + "learning_rate": 1.2671025572746353e-07, + "loss": 0.5459, + "step": 3914 + }, + { + "epoch": 0.9305918706917043, + "grad_norm": 1.122619084095432, + "learning_rate": 1.2585045584791623e-07, + "loss": 0.5351, + "step": 3915 + }, + { + "epoch": 0.9308295697646779, + "grad_norm": 1.2050885519394887, + "learning_rate": 1.2499354583797473e-07, + "loss": 0.5297, + "step": 3916 + }, + { + "epoch": 0.9310672688376516, + "grad_norm": 1.1897646192128102, + "learning_rate": 1.241395262056999e-07, + "loss": 0.5086, + "step": 3917 + }, + { + "epoch": 0.9313049679106251, + "grad_norm": 1.2168659551900671, + "learning_rate": 1.232883974574367e-07, + "loss": 0.5781, + "step": 3918 + }, + { + "epoch": 0.9315426669835988, + "grad_norm": 1.1800891455066036, + "learning_rate": 1.22440160097817e-07, + "loss": 0.5413, + "step": 3919 + }, + { + "epoch": 0.9317803660565723, + "grad_norm": 1.2009339245330974, + "learning_rate": 1.2159481462975797e-07, + "loss": 0.5215, + "step": 3920 + }, + { + "epoch": 0.932018065129546, + "grad_norm": 1.2512648223847822, + "learning_rate": 1.2075236155446423e-07, + "loss": 0.5198, + "step": 3921 + }, + { + "epoch": 0.9322557642025197, + "grad_norm": 1.1587631576740045, + "learning_rate": 1.199128013714218e-07, + "loss": 0.5163, + "step": 3922 + }, + { + "epoch": 0.9324934632754932, + "grad_norm": 1.2314704610573048, + "learning_rate": 1.1907613457840472e-07, + "loss": 0.5685, + "step": 3923 + }, + { + "epoch": 0.9327311623484669, + "grad_norm": 1.881665558581674, + "learning_rate": 1.1824236167146952e-07, + "loss": 0.5316, + "step": 3924 + }, + { + "epoch": 0.9329688614214404, + "grad_norm": 1.2781181305853535, + "learning_rate": 1.1741148314495965e-07, + "loss": 0.5368, + "step": 3925 + }, + { + "epoch": 0.9332065604944141, + "grad_norm": 1.2136690388184448, + "learning_rate": 1.1658349949149882e-07, + "loss": 0.5328, + "step": 3926 + }, + { + "epoch": 0.9334442595673876, + "grad_norm": 1.3510200326656807, + "learning_rate": 1.157584112019966e-07, + "loss": 0.4903, + "step": 3927 + }, + { + "epoch": 0.9336819586403613, + "grad_norm": 1.4588900897610841, + "learning_rate": 1.1493621876564554e-07, + "loss": 0.536, + "step": 3928 + }, + { + "epoch": 0.933919657713335, + "grad_norm": 1.1223772809958958, + "learning_rate": 1.1411692266992237e-07, + "loss": 0.5251, + "step": 3929 + }, + { + "epoch": 0.9341573567863085, + "grad_norm": 1.1879640469407127, + "learning_rate": 1.133005234005835e-07, + "loss": 0.5503, + "step": 3930 + }, + { + "epoch": 0.9343950558592822, + "grad_norm": 1.2595600017965953, + "learning_rate": 1.1248702144167123e-07, + "loss": 0.545, + "step": 3931 + }, + { + "epoch": 0.9346327549322557, + "grad_norm": 1.4388136189911351, + "learning_rate": 1.1167641727550804e-07, + "loss": 0.5479, + "step": 3932 + }, + { + "epoch": 0.9348704540052294, + "grad_norm": 1.347875583344727, + "learning_rate": 1.1086871138270006e-07, + "loss": 0.5658, + "step": 3933 + }, + { + "epoch": 0.9351081530782029, + "grad_norm": 1.189709842225491, + "learning_rate": 1.1006390424213143e-07, + "loss": 0.568, + "step": 3934 + }, + { + "epoch": 0.9353458521511766, + "grad_norm": 1.2070027992599839, + "learning_rate": 1.0926199633097156e-07, + "loss": 0.619, + "step": 3935 + }, + { + "epoch": 0.9355835512241503, + "grad_norm": 1.1892869546344929, + "learning_rate": 1.084629881246696e-07, + "loss": 0.4978, + "step": 3936 + }, + { + "epoch": 0.9358212502971238, + "grad_norm": 1.1018025348665321, + "learning_rate": 1.0766688009695548e-07, + "loss": 0.4975, + "step": 3937 + }, + { + "epoch": 0.9360589493700975, + "grad_norm": 1.175285658592407, + "learning_rate": 1.0687367271983772e-07, + "loss": 0.5763, + "step": 3938 + }, + { + "epoch": 0.936296648443071, + "grad_norm": 1.318201669134416, + "learning_rate": 1.060833664636085e-07, + "loss": 0.5223, + "step": 3939 + }, + { + "epoch": 0.9365343475160447, + "grad_norm": 1.3176999175562796, + "learning_rate": 1.0529596179683743e-07, + "loss": 0.5508, + "step": 3940 + }, + { + "epoch": 0.9367720465890182, + "grad_norm": 1.207802300518822, + "learning_rate": 1.0451145918637495e-07, + "loss": 0.5623, + "step": 3941 + }, + { + "epoch": 0.9370097456619919, + "grad_norm": 1.6101958913034862, + "learning_rate": 1.0372985909734956e-07, + "loss": 0.5678, + "step": 3942 + }, + { + "epoch": 0.9372474447349656, + "grad_norm": 1.2389033263959255, + "learning_rate": 1.0295116199317057e-07, + "loss": 0.4739, + "step": 3943 + }, + { + "epoch": 0.9374851438079391, + "grad_norm": 1.2492352713667771, + "learning_rate": 1.0217536833552533e-07, + "loss": 0.5909, + "step": 3944 + }, + { + "epoch": 0.9377228428809128, + "grad_norm": 1.652140116716469, + "learning_rate": 1.0140247858437979e-07, + "loss": 0.5302, + "step": 3945 + }, + { + "epoch": 0.9379605419538863, + "grad_norm": 1.3494990145505548, + "learning_rate": 1.0063249319797741e-07, + "loss": 0.5575, + "step": 3946 + }, + { + "epoch": 0.93819824102686, + "grad_norm": 1.2337014798017685, + "learning_rate": 9.986541263284077e-08, + "loss": 0.5516, + "step": 3947 + }, + { + "epoch": 0.9384359400998337, + "grad_norm": 1.2177641756828002, + "learning_rate": 9.910123734376997e-08, + "loss": 0.5563, + "step": 3948 + }, + { + "epoch": 0.9386736391728072, + "grad_norm": 1.2902437289081161, + "learning_rate": 9.833996778384259e-08, + "loss": 0.5585, + "step": 3949 + }, + { + "epoch": 0.9389113382457809, + "grad_norm": 1.2970818997446938, + "learning_rate": 9.7581604404412e-08, + "loss": 0.5345, + "step": 3950 + }, + { + "epoch": 0.9391490373187544, + "grad_norm": 1.3496272069921023, + "learning_rate": 9.682614765511134e-08, + "loss": 0.5377, + "step": 3951 + }, + { + "epoch": 0.9393867363917281, + "grad_norm": 1.2607353810079003, + "learning_rate": 9.607359798384785e-08, + "loss": 0.5538, + "step": 3952 + }, + { + "epoch": 0.9396244354647016, + "grad_norm": 1.4014254896569631, + "learning_rate": 9.532395583680632e-08, + "loss": 0.4902, + "step": 3953 + }, + { + "epoch": 0.9398621345376753, + "grad_norm": 1.5403867952537185, + "learning_rate": 9.457722165844729e-08, + "loss": 0.5178, + "step": 3954 + }, + { + "epoch": 0.940099833610649, + "grad_norm": 2.134160949377203, + "learning_rate": 9.383339589150776e-08, + "loss": 0.5751, + "step": 3955 + }, + { + "epoch": 0.9403375326836225, + "grad_norm": 1.279635509523856, + "learning_rate": 9.309247897700046e-08, + "loss": 0.5457, + "step": 3956 + }, + { + "epoch": 0.9405752317565962, + "grad_norm": 1.1805589397861573, + "learning_rate": 9.235447135421127e-08, + "loss": 0.5051, + "step": 3957 + }, + { + "epoch": 0.9408129308295697, + "grad_norm": 1.1776923219804265, + "learning_rate": 9.161937346070404e-08, + "loss": 0.5451, + "step": 3958 + }, + { + "epoch": 0.9410506299025434, + "grad_norm": 1.6161225617746335, + "learning_rate": 9.08871857323157e-08, + "loss": 0.5404, + "step": 3959 + }, + { + "epoch": 0.941288328975517, + "grad_norm": 1.473238376071312, + "learning_rate": 9.015790860315954e-08, + "loss": 0.5665, + "step": 3960 + }, + { + "epoch": 0.9415260280484906, + "grad_norm": 1.2661308026110314, + "learning_rate": 8.943154250562025e-08, + "loss": 0.5321, + "step": 3961 + }, + { + "epoch": 0.9417637271214643, + "grad_norm": 1.1323837599011128, + "learning_rate": 8.87080878703589e-08, + "loss": 0.5367, + "step": 3962 + }, + { + "epoch": 0.9420014261944378, + "grad_norm": 1.1729660235909614, + "learning_rate": 8.798754512631013e-08, + "loss": 0.5248, + "step": 3963 + }, + { + "epoch": 0.9422391252674115, + "grad_norm": 1.1803248807764684, + "learning_rate": 8.726991470068169e-08, + "loss": 0.5338, + "step": 3964 + }, + { + "epoch": 0.942476824340385, + "grad_norm": 1.2191822313887266, + "learning_rate": 8.655519701895376e-08, + "loss": 0.5982, + "step": 3965 + }, + { + "epoch": 0.9427145234133587, + "grad_norm": 1.5100343506402374, + "learning_rate": 8.584339250488127e-08, + "loss": 0.5478, + "step": 3966 + }, + { + "epoch": 0.9429522224863323, + "grad_norm": 1.6182896431335971, + "learning_rate": 8.513450158049109e-08, + "loss": 0.5729, + "step": 3967 + }, + { + "epoch": 0.9431899215593059, + "grad_norm": 1.2238643224622552, + "learning_rate": 8.442852466608253e-08, + "loss": 0.5612, + "step": 3968 + }, + { + "epoch": 0.9434276206322796, + "grad_norm": 1.3325427456054613, + "learning_rate": 8.372546218022747e-08, + "loss": 0.572, + "step": 3969 + }, + { + "epoch": 0.9436653197052531, + "grad_norm": 1.3031500908910345, + "learning_rate": 8.302531453976914e-08, + "loss": 0.5736, + "step": 3970 + }, + { + "epoch": 0.9439030187782268, + "grad_norm": 1.3612537518050039, + "learning_rate": 8.232808215982325e-08, + "loss": 0.5292, + "step": 3971 + }, + { + "epoch": 0.9441407178512004, + "grad_norm": 1.166780167789163, + "learning_rate": 8.163376545377744e-08, + "loss": 0.5062, + "step": 3972 + }, + { + "epoch": 0.944378416924174, + "grad_norm": 1.4569853889994027, + "learning_rate": 8.094236483329022e-08, + "loss": 0.5475, + "step": 3973 + }, + { + "epoch": 0.9446161159971476, + "grad_norm": 1.493230157557986, + "learning_rate": 8.025388070828977e-08, + "loss": 0.5382, + "step": 3974 + }, + { + "epoch": 0.9448538150701212, + "grad_norm": 1.2350104600552656, + "learning_rate": 7.956831348697791e-08, + "loss": 0.5167, + "step": 3975 + }, + { + "epoch": 0.9450915141430949, + "grad_norm": 1.235408188414576, + "learning_rate": 7.888566357582506e-08, + "loss": 0.5278, + "step": 3976 + }, + { + "epoch": 0.9453292132160684, + "grad_norm": 1.1169352639689132, + "learning_rate": 7.820593137957244e-08, + "loss": 0.5304, + "step": 3977 + }, + { + "epoch": 0.9455669122890421, + "grad_norm": 1.147296997193163, + "learning_rate": 7.752911730123102e-08, + "loss": 0.524, + "step": 3978 + }, + { + "epoch": 0.9458046113620157, + "grad_norm": 1.1742306121690722, + "learning_rate": 7.685522174208205e-08, + "loss": 0.5397, + "step": 3979 + }, + { + "epoch": 0.9460423104349893, + "grad_norm": 1.3303068003229324, + "learning_rate": 7.618424510167644e-08, + "loss": 0.5616, + "step": 3980 + }, + { + "epoch": 0.9462800095079629, + "grad_norm": 1.2547471691541858, + "learning_rate": 7.551618777783487e-08, + "loss": 0.5463, + "step": 3981 + }, + { + "epoch": 0.9465177085809365, + "grad_norm": 1.3049533966217577, + "learning_rate": 7.485105016664551e-08, + "loss": 0.5701, + "step": 3982 + }, + { + "epoch": 0.9467554076539102, + "grad_norm": 1.4326836160830199, + "learning_rate": 7.418883266246734e-08, + "loss": 0.5603, + "step": 3983 + }, + { + "epoch": 0.9469931067268837, + "grad_norm": 1.3056787772610043, + "learning_rate": 7.352953565792798e-08, + "loss": 0.5179, + "step": 3984 + }, + { + "epoch": 0.9472308057998574, + "grad_norm": 1.2918821524627084, + "learning_rate": 7.287315954392137e-08, + "loss": 0.5742, + "step": 3985 + }, + { + "epoch": 0.947468504872831, + "grad_norm": 1.2432919516618277, + "learning_rate": 7.221970470961125e-08, + "loss": 0.5452, + "step": 3986 + }, + { + "epoch": 0.9477062039458046, + "grad_norm": 1.5596310872883354, + "learning_rate": 7.156917154243048e-08, + "loss": 0.5679, + "step": 3987 + }, + { + "epoch": 0.9479439030187782, + "grad_norm": 1.1482653130634906, + "learning_rate": 7.092156042807719e-08, + "loss": 0.5031, + "step": 3988 + }, + { + "epoch": 0.9481816020917518, + "grad_norm": 1.2079438156006226, + "learning_rate": 7.027687175051812e-08, + "loss": 0.5701, + "step": 3989 + }, + { + "epoch": 0.9484193011647255, + "grad_norm": 1.3744698771349015, + "learning_rate": 6.963510589198753e-08, + "loss": 0.4908, + "step": 3990 + }, + { + "epoch": 0.948657000237699, + "grad_norm": 1.1330438857979523, + "learning_rate": 6.899626323298714e-08, + "loss": 0.5028, + "step": 3991 + }, + { + "epoch": 0.9488946993106727, + "grad_norm": 1.118984865549763, + "learning_rate": 6.836034415228509e-08, + "loss": 0.5363, + "step": 3992 + }, + { + "epoch": 0.9491323983836463, + "grad_norm": 1.2291110402336667, + "learning_rate": 6.77273490269148e-08, + "loss": 0.5252, + "step": 3993 + }, + { + "epoch": 0.9493700974566199, + "grad_norm": 1.3126223357843652, + "learning_rate": 6.709727823217827e-08, + "loss": 0.5216, + "step": 3994 + }, + { + "epoch": 0.9496077965295935, + "grad_norm": 1.1803472178234076, + "learning_rate": 6.647013214164277e-08, + "loss": 0.5196, + "step": 3995 + }, + { + "epoch": 0.9498454956025671, + "grad_norm": 1.1675636465361665, + "learning_rate": 6.584591112714089e-08, + "loss": 0.538, + "step": 3996 + }, + { + "epoch": 0.9500831946755408, + "grad_norm": 1.185979711977627, + "learning_rate": 6.522461555877213e-08, + "loss": 0.5343, + "step": 3997 + }, + { + "epoch": 0.9503208937485144, + "grad_norm": 1.445299243503812, + "learning_rate": 6.460624580490016e-08, + "loss": 0.5589, + "step": 3998 + }, + { + "epoch": 0.950558592821488, + "grad_norm": 1.305742737667683, + "learning_rate": 6.399080223215503e-08, + "loss": 0.5528, + "step": 3999 + }, + { + "epoch": 0.9507962918944616, + "grad_norm": 1.2928322812259199, + "learning_rate": 6.33782852054321e-08, + "loss": 0.5381, + "step": 4000 + }, + { + "epoch": 0.9507962918944616, + "eval_loss": 0.5519807934761047, + "eval_runtime": 285.4841, + "eval_samples_per_second": 38.104, + "eval_steps_per_second": 1.191, + "step": 4000 + }, + { + "epoch": 0.9510339909674352, + "grad_norm": 1.5526466560504188, + "learning_rate": 6.27686950878903e-08, + "loss": 0.5355, + "step": 4001 + }, + { + "epoch": 0.9512716900404088, + "grad_norm": 1.3613766709132775, + "learning_rate": 6.216203224095386e-08, + "loss": 0.5603, + "step": 4002 + }, + { + "epoch": 0.9515093891133825, + "grad_norm": 1.2712552495039617, + "learning_rate": 6.15582970243117e-08, + "loss": 0.5496, + "step": 4003 + }, + { + "epoch": 0.9517470881863561, + "grad_norm": 1.2072289804095282, + "learning_rate": 6.095748979591698e-08, + "loss": 0.5523, + "step": 4004 + }, + { + "epoch": 0.9519847872593297, + "grad_norm": 1.3641183949211353, + "learning_rate": 6.035961091198584e-08, + "loss": 0.5463, + "step": 4005 + }, + { + "epoch": 0.9522224863323033, + "grad_norm": 1.2569092908737456, + "learning_rate": 5.976466072699971e-08, + "loss": 0.5353, + "step": 4006 + }, + { + "epoch": 0.9524601854052769, + "grad_norm": 1.344170085561234, + "learning_rate": 5.917263959370312e-08, + "loss": 0.5231, + "step": 4007 + }, + { + "epoch": 0.9526978844782505, + "grad_norm": 1.2269476311768264, + "learning_rate": 5.858354786310305e-08, + "loss": 0.481, + "step": 4008 + }, + { + "epoch": 0.9529355835512241, + "grad_norm": 1.2699191092688447, + "learning_rate": 5.799738588447068e-08, + "loss": 0.5611, + "step": 4009 + }, + { + "epoch": 0.9531732826241978, + "grad_norm": 1.212320626581182, + "learning_rate": 5.741415400533967e-08, + "loss": 0.5201, + "step": 4010 + }, + { + "epoch": 0.9534109816971714, + "grad_norm": 1.2614832687852655, + "learning_rate": 5.683385257150675e-08, + "loss": 0.5982, + "step": 4011 + }, + { + "epoch": 0.953648680770145, + "grad_norm": 1.2627597395405095, + "learning_rate": 5.625648192703115e-08, + "loss": 0.5344, + "step": 4012 + }, + { + "epoch": 0.9538863798431186, + "grad_norm": 1.1461878213131136, + "learning_rate": 5.568204241423347e-08, + "loss": 0.5169, + "step": 4013 + }, + { + "epoch": 0.9541240789160922, + "grad_norm": 1.3492956975378008, + "learning_rate": 5.511053437369795e-08, + "loss": 0.5536, + "step": 4014 + }, + { + "epoch": 0.9543617779890659, + "grad_norm": 1.1453469497559075, + "learning_rate": 5.454195814427021e-08, + "loss": 0.5722, + "step": 4015 + }, + { + "epoch": 0.9545994770620394, + "grad_norm": 1.2578051310231229, + "learning_rate": 5.3976314063057256e-08, + "loss": 0.5408, + "step": 4016 + }, + { + "epoch": 0.9548371761350131, + "grad_norm": 1.2528852488751874, + "learning_rate": 5.341360246542804e-08, + "loss": 0.5428, + "step": 4017 + }, + { + "epoch": 0.9550748752079867, + "grad_norm": 1.5162742055513845, + "learning_rate": 5.285382368501235e-08, + "loss": 0.544, + "step": 4018 + }, + { + "epoch": 0.9553125742809603, + "grad_norm": 1.1988090923455283, + "learning_rate": 5.2296978053702486e-08, + "loss": 0.5539, + "step": 4019 + }, + { + "epoch": 0.9555502733539339, + "grad_norm": 1.6784742312658107, + "learning_rate": 5.174306590164879e-08, + "loss": 0.5559, + "step": 4020 + }, + { + "epoch": 0.9557879724269075, + "grad_norm": 1.212854493525016, + "learning_rate": 5.119208755726579e-08, + "loss": 0.5501, + "step": 4021 + }, + { + "epoch": 0.9560256714998812, + "grad_norm": 1.253151534468677, + "learning_rate": 5.0644043347226615e-08, + "loss": 0.547, + "step": 4022 + }, + { + "epoch": 0.9562633705728548, + "grad_norm": 1.235125078593384, + "learning_rate": 5.009893359646523e-08, + "loss": 0.5247, + "step": 4023 + }, + { + "epoch": 0.9565010696458284, + "grad_norm": 1.137741421978973, + "learning_rate": 4.955675862817533e-08, + "loss": 0.5299, + "step": 4024 + }, + { + "epoch": 0.956738768718802, + "grad_norm": 1.4067942445909405, + "learning_rate": 4.901751876381089e-08, + "loss": 0.5069, + "step": 4025 + }, + { + "epoch": 0.9569764677917756, + "grad_norm": 1.2037292447277643, + "learning_rate": 4.848121432308617e-08, + "loss": 0.4871, + "step": 4026 + }, + { + "epoch": 0.9572141668647492, + "grad_norm": 1.1780580990303342, + "learning_rate": 4.794784562397459e-08, + "loss": 0.5592, + "step": 4027 + }, + { + "epoch": 0.9574518659377228, + "grad_norm": 1.1576520242026505, + "learning_rate": 4.741741298270819e-08, + "loss": 0.5358, + "step": 4028 + }, + { + "epoch": 0.9576895650106965, + "grad_norm": 1.2837473622925062, + "learning_rate": 4.688991671377929e-08, + "loss": 0.5484, + "step": 4029 + }, + { + "epoch": 0.9579272640836701, + "grad_norm": 1.271134578514592, + "learning_rate": 4.636535712993939e-08, + "loss": 0.5172, + "step": 4030 + }, + { + "epoch": 0.9581649631566437, + "grad_norm": 1.2689065757037583, + "learning_rate": 4.584373454219859e-08, + "loss": 0.5359, + "step": 4031 + }, + { + "epoch": 0.9584026622296173, + "grad_norm": 1.2004387870845241, + "learning_rate": 4.532504925982506e-08, + "loss": 0.5346, + "step": 4032 + }, + { + "epoch": 0.9586403613025909, + "grad_norm": 1.248966372574798, + "learning_rate": 4.4809301590345576e-08, + "loss": 0.5091, + "step": 4033 + }, + { + "epoch": 0.9588780603755646, + "grad_norm": 1.2554495067412024, + "learning_rate": 4.429649183954554e-08, + "loss": 0.5208, + "step": 4034 + }, + { + "epoch": 0.9591157594485381, + "grad_norm": 1.1633224879630701, + "learning_rate": 4.378662031146896e-08, + "loss": 0.5777, + "step": 4035 + }, + { + "epoch": 0.9593534585215118, + "grad_norm": 1.2134740079026276, + "learning_rate": 4.327968730841681e-08, + "loss": 0.5435, + "step": 4036 + }, + { + "epoch": 0.9595911575944854, + "grad_norm": 1.2280502151822943, + "learning_rate": 4.2775693130948094e-08, + "loss": 0.5645, + "step": 4037 + }, + { + "epoch": 0.959828856667459, + "grad_norm": 1.127410663601608, + "learning_rate": 4.2274638077879905e-08, + "loss": 0.5702, + "step": 4038 + }, + { + "epoch": 0.9600665557404326, + "grad_norm": 1.2848925122811181, + "learning_rate": 4.177652244628627e-08, + "loss": 0.4968, + "step": 4039 + }, + { + "epoch": 0.9603042548134062, + "grad_norm": 1.2001817470074512, + "learning_rate": 4.128134653149874e-08, + "loss": 0.5196, + "step": 4040 + }, + { + "epoch": 0.9605419538863799, + "grad_norm": 1.3073664560671268, + "learning_rate": 4.078911062710522e-08, + "loss": 0.551, + "step": 4041 + }, + { + "epoch": 0.9607796529593534, + "grad_norm": 1.1041518409587907, + "learning_rate": 4.029981502495117e-08, + "loss": 0.5085, + "step": 4042 + }, + { + "epoch": 0.9610173520323271, + "grad_norm": 1.2414512564844453, + "learning_rate": 3.9813460015139524e-08, + "loss": 0.5146, + "step": 4043 + }, + { + "epoch": 0.9612550511053007, + "grad_norm": 1.2435541770945768, + "learning_rate": 3.93300458860274e-08, + "loss": 0.4887, + "step": 4044 + }, + { + "epoch": 0.9614927501782743, + "grad_norm": 1.403452677517591, + "learning_rate": 3.884957292422997e-08, + "loss": 0.5549, + "step": 4045 + }, + { + "epoch": 0.961730449251248, + "grad_norm": 1.178670387262961, + "learning_rate": 3.837204141461881e-08, + "loss": 0.5295, + "step": 4046 + }, + { + "epoch": 0.9619681483242215, + "grad_norm": 1.2926561631480191, + "learning_rate": 3.7897451640321326e-08, + "loss": 0.5552, + "step": 4047 + }, + { + "epoch": 0.9622058473971952, + "grad_norm": 1.1295119130029614, + "learning_rate": 3.742580388271966e-08, + "loss": 0.5455, + "step": 4048 + }, + { + "epoch": 0.9624435464701687, + "grad_norm": 1.5427508377328605, + "learning_rate": 3.695709842145234e-08, + "loss": 0.5951, + "step": 4049 + }, + { + "epoch": 0.9626812455431424, + "grad_norm": 1.160925204851751, + "learning_rate": 3.6491335534414305e-08, + "loss": 0.4978, + "step": 4050 + }, + { + "epoch": 0.962918944616116, + "grad_norm": 1.2961045472705186, + "learning_rate": 3.602851549775521e-08, + "loss": 0.6035, + "step": 4051 + }, + { + "epoch": 0.9631566436890896, + "grad_norm": 1.2561770244730086, + "learning_rate": 3.556863858587833e-08, + "loss": 0.5087, + "step": 4052 + }, + { + "epoch": 0.9633943427620633, + "grad_norm": 1.1801598646835594, + "learning_rate": 3.5111705071444455e-08, + "loss": 0.5952, + "step": 4053 + }, + { + "epoch": 0.9636320418350368, + "grad_norm": 1.2149225822448393, + "learning_rate": 3.465771522536854e-08, + "loss": 0.5842, + "step": 4054 + }, + { + "epoch": 0.9638697409080105, + "grad_norm": 1.2516130350267345, + "learning_rate": 3.4206669316819155e-08, + "loss": 0.5422, + "step": 4055 + }, + { + "epoch": 0.964107439980984, + "grad_norm": 1.274117875435793, + "learning_rate": 3.375856761322016e-08, + "loss": 0.5125, + "step": 4056 + }, + { + "epoch": 0.9643451390539577, + "grad_norm": 2.1020985952525972, + "learning_rate": 3.3313410380250157e-08, + "loss": 0.5259, + "step": 4057 + }, + { + "epoch": 0.9645828381269314, + "grad_norm": 1.3759509875730533, + "learning_rate": 3.2871197881841346e-08, + "loss": 0.508, + "step": 4058 + }, + { + "epoch": 0.9648205371999049, + "grad_norm": 1.5574704812902664, + "learning_rate": 3.243193038018122e-08, + "loss": 0.5025, + "step": 4059 + }, + { + "epoch": 0.9650582362728786, + "grad_norm": 1.2914566471580076, + "learning_rate": 3.199560813570868e-08, + "loss": 0.5522, + "step": 4060 + }, + { + "epoch": 0.9652959353458521, + "grad_norm": 1.591702559666577, + "learning_rate": 3.156223140711845e-08, + "loss": 0.5603, + "step": 4061 + }, + { + "epoch": 0.9655336344188258, + "grad_norm": 1.2066171943139363, + "learning_rate": 3.113180045135944e-08, + "loss": 0.5378, + "step": 4062 + }, + { + "epoch": 0.9657713334917993, + "grad_norm": 2.2300181693120953, + "learning_rate": 3.0704315523631956e-08, + "loss": 0.5361, + "step": 4063 + }, + { + "epoch": 0.966009032564773, + "grad_norm": 1.2784751183765233, + "learning_rate": 3.0279776877390474e-08, + "loss": 0.5216, + "step": 4064 + }, + { + "epoch": 0.9662467316377467, + "grad_norm": 1.3548480636925493, + "learning_rate": 2.9858184764343104e-08, + "loss": 0.5083, + "step": 4065 + }, + { + "epoch": 0.9664844307107202, + "grad_norm": 1.2303328779900766, + "learning_rate": 2.9439539434450994e-08, + "loss": 0.499, + "step": 4066 + }, + { + "epoch": 0.9667221297836939, + "grad_norm": 1.2932568763074626, + "learning_rate": 2.9023841135927822e-08, + "loss": 0.5247, + "step": 4067 + }, + { + "epoch": 0.9669598288566674, + "grad_norm": 1.5073555716096394, + "learning_rate": 2.861109011523866e-08, + "loss": 0.5465, + "step": 4068 + }, + { + "epoch": 0.9671975279296411, + "grad_norm": 1.417244142744646, + "learning_rate": 2.8201286617103863e-08, + "loss": 0.5768, + "step": 4069 + }, + { + "epoch": 0.9674352270026146, + "grad_norm": 1.200398607232495, + "learning_rate": 2.7794430884494073e-08, + "loss": 0.5111, + "step": 4070 + }, + { + "epoch": 0.9676729260755883, + "grad_norm": 1.3175303321567582, + "learning_rate": 2.7390523158633552e-08, + "loss": 0.5603, + "step": 4071 + }, + { + "epoch": 0.967910625148562, + "grad_norm": 1.1577780714185708, + "learning_rate": 2.6989563678996856e-08, + "loss": 0.5203, + "step": 4072 + }, + { + "epoch": 0.9681483242215355, + "grad_norm": 1.2941088367241917, + "learning_rate": 2.659155268331326e-08, + "loss": 0.5748, + "step": 4073 + }, + { + "epoch": 0.9683860232945092, + "grad_norm": 1.3252053264394446, + "learning_rate": 2.6196490407561227e-08, + "loss": 0.5165, + "step": 4074 + }, + { + "epoch": 0.9686237223674827, + "grad_norm": 1.1623300765621767, + "learning_rate": 2.5804377085972278e-08, + "loss": 0.512, + "step": 4075 + }, + { + "epoch": 0.9688614214404564, + "grad_norm": 1.0906754420015778, + "learning_rate": 2.5415212951029334e-08, + "loss": 0.4966, + "step": 4076 + }, + { + "epoch": 0.96909912051343, + "grad_norm": 1.235842498638525, + "learning_rate": 2.5028998233467272e-08, + "loss": 0.4832, + "step": 4077 + }, + { + "epoch": 0.9693368195864036, + "grad_norm": 1.2181781866645416, + "learning_rate": 2.4645733162271255e-08, + "loss": 0.563, + "step": 4078 + }, + { + "epoch": 0.9695745186593773, + "grad_norm": 1.388346840937979, + "learning_rate": 2.426541796467785e-08, + "loss": 0.5165, + "step": 4079 + }, + { + "epoch": 0.9698122177323508, + "grad_norm": 1.1467340088787639, + "learning_rate": 2.388805286617557e-08, + "loss": 0.5355, + "step": 4080 + }, + { + "epoch": 0.9700499168053245, + "grad_norm": 1.1917745850346888, + "learning_rate": 2.351363809050211e-08, + "loss": 0.5779, + "step": 4081 + }, + { + "epoch": 0.970287615878298, + "grad_norm": 1.0993209581197656, + "learning_rate": 2.3142173859647675e-08, + "loss": 0.5419, + "step": 4082 + }, + { + "epoch": 0.9705253149512717, + "grad_norm": 1.3603702976429224, + "learning_rate": 2.2773660393852758e-08, + "loss": 0.5593, + "step": 4083 + }, + { + "epoch": 0.9707630140242453, + "grad_norm": 1.3787909849048285, + "learning_rate": 2.2408097911606473e-08, + "loss": 0.485, + "step": 4084 + }, + { + "epoch": 0.9710007130972189, + "grad_norm": 1.0946626174565905, + "learning_rate": 2.204548662965156e-08, + "loss": 0.5292, + "step": 4085 + }, + { + "epoch": 0.9712384121701926, + "grad_norm": 1.3193308494718874, + "learning_rate": 2.1685826762977703e-08, + "loss": 0.5207, + "step": 4086 + }, + { + "epoch": 0.9714761112431661, + "grad_norm": 1.279145237949268, + "learning_rate": 2.1329118524827662e-08, + "loss": 0.5481, + "step": 4087 + }, + { + "epoch": 0.9717138103161398, + "grad_norm": 1.401254722257118, + "learning_rate": 2.097536212669171e-08, + "loss": 0.5524, + "step": 4088 + }, + { + "epoch": 0.9719515093891133, + "grad_norm": 1.380521192482498, + "learning_rate": 2.062455777831096e-08, + "loss": 0.5396, + "step": 4089 + }, + { + "epoch": 0.972189208462087, + "grad_norm": 1.1940025606337061, + "learning_rate": 2.0276705687676813e-08, + "loss": 0.5394, + "step": 4090 + }, + { + "epoch": 0.9724269075350606, + "grad_norm": 1.3589158335989469, + "learning_rate": 1.9931806061029846e-08, + "loss": 0.5359, + "step": 4091 + }, + { + "epoch": 0.9726646066080342, + "grad_norm": 1.151835044880591, + "learning_rate": 1.958985910285982e-08, + "loss": 0.5188, + "step": 4092 + }, + { + "epoch": 0.9729023056810079, + "grad_norm": 1.2917842315868864, + "learning_rate": 1.9250865015906784e-08, + "loss": 0.5779, + "step": 4093 + }, + { + "epoch": 0.9731400047539814, + "grad_norm": 1.1061179640851506, + "learning_rate": 1.8914824001158295e-08, + "loss": 0.5623, + "step": 4094 + }, + { + "epoch": 0.9733777038269551, + "grad_norm": 1.4214644120075035, + "learning_rate": 1.8581736257852756e-08, + "loss": 0.5933, + "step": 4095 + }, + { + "epoch": 0.9736154028999287, + "grad_norm": 1.4806811808327742, + "learning_rate": 1.825160198347664e-08, + "loss": 0.5421, + "step": 4096 + }, + { + "epoch": 0.9738531019729023, + "grad_norm": 1.3724752509230207, + "learning_rate": 1.7924421373766153e-08, + "loss": 0.5731, + "step": 4097 + }, + { + "epoch": 0.974090801045876, + "grad_norm": 1.187227790073092, + "learning_rate": 1.7600194622705014e-08, + "loss": 0.5372, + "step": 4098 + }, + { + "epoch": 0.9743285001188495, + "grad_norm": 1.3219440679250685, + "learning_rate": 1.7278921922527224e-08, + "loss": 0.5344, + "step": 4099 + }, + { + "epoch": 0.9745661991918232, + "grad_norm": 1.2550450466456873, + "learning_rate": 1.696060346371431e-08, + "loss": 0.5653, + "step": 4100 + }, + { + "epoch": 0.9745661991918232, + "eval_loss": 0.5519022345542908, + "eval_runtime": 290.6047, + "eval_samples_per_second": 37.432, + "eval_steps_per_second": 1.17, + "step": 4100 + }, + { + "epoch": 0.9748038982647967, + "grad_norm": 1.2721040968120674, + "learning_rate": 1.6645239434995297e-08, + "loss": 0.5519, + "step": 4101 + }, + { + "epoch": 0.9750415973377704, + "grad_norm": 1.195335668078204, + "learning_rate": 1.6332830023350065e-08, + "loss": 0.5445, + "step": 4102 + }, + { + "epoch": 0.975279296410744, + "grad_norm": 1.1292808242448638, + "learning_rate": 1.6023375414004894e-08, + "loss": 0.489, + "step": 4103 + }, + { + "epoch": 0.9755169954837176, + "grad_norm": 1.3832345368099073, + "learning_rate": 1.571687579043357e-08, + "loss": 0.5457, + "step": 4104 + }, + { + "epoch": 0.9757546945566913, + "grad_norm": 1.1289730338023682, + "learning_rate": 1.541333133436018e-08, + "loss": 0.4962, + "step": 4105 + }, + { + "epoch": 0.9759923936296648, + "grad_norm": 1.1912471919588388, + "learning_rate": 1.511274222575465e-08, + "loss": 0.505, + "step": 4106 + }, + { + "epoch": 0.9762300927026385, + "grad_norm": 1.319966607037857, + "learning_rate": 1.481510864283553e-08, + "loss": 0.5416, + "step": 4107 + }, + { + "epoch": 0.976467791775612, + "grad_norm": 1.2404772136709803, + "learning_rate": 1.4520430762069438e-08, + "loss": 0.489, + "step": 4108 + }, + { + "epoch": 0.9767054908485857, + "grad_norm": 1.3874787236362027, + "learning_rate": 1.4228708758169396e-08, + "loss": 0.5047, + "step": 4109 + }, + { + "epoch": 0.9769431899215593, + "grad_norm": 1.314660854205586, + "learning_rate": 1.3939942804096496e-08, + "loss": 0.5707, + "step": 4110 + }, + { + "epoch": 0.9771808889945329, + "grad_norm": 1.3931237693933662, + "learning_rate": 1.3654133071059894e-08, + "loss": 0.5602, + "step": 4111 + }, + { + "epoch": 0.9774185880675066, + "grad_norm": 1.1667459263137545, + "learning_rate": 1.3371279728515152e-08, + "loss": 0.5259, + "step": 4112 + }, + { + "epoch": 0.9776562871404801, + "grad_norm": 1.2507876624663952, + "learning_rate": 1.3091382944164788e-08, + "loss": 0.5169, + "step": 4113 + }, + { + "epoch": 0.9778939862134538, + "grad_norm": 1.565928116296441, + "learning_rate": 1.2814442883959388e-08, + "loss": 0.558, + "step": 4114 + }, + { + "epoch": 0.9781316852864274, + "grad_norm": 1.2629233938714097, + "learning_rate": 1.2540459712095387e-08, + "loss": 0.5313, + "step": 4115 + }, + { + "epoch": 0.978369384359401, + "grad_norm": 1.7348571004010722, + "learning_rate": 1.2269433591017288e-08, + "loss": 0.5171, + "step": 4116 + }, + { + "epoch": 0.9786070834323746, + "grad_norm": 1.190160229101479, + "learning_rate": 1.200136468141544e-08, + "loss": 0.5484, + "step": 4117 + }, + { + "epoch": 0.9788447825053482, + "grad_norm": 1.6329549288610559, + "learning_rate": 1.1736253142226594e-08, + "loss": 0.5825, + "step": 4118 + }, + { + "epoch": 0.9790824815783219, + "grad_norm": 1.440793516344676, + "learning_rate": 1.1474099130635575e-08, + "loss": 0.5303, + "step": 4119 + }, + { + "epoch": 0.9793201806512954, + "grad_norm": 1.646961602032492, + "learning_rate": 1.1214902802072491e-08, + "loss": 0.5224, + "step": 4120 + }, + { + "epoch": 0.9795578797242691, + "grad_norm": 1.2460014504879438, + "learning_rate": 1.0958664310213862e-08, + "loss": 0.5694, + "step": 4121 + }, + { + "epoch": 0.9797955787972427, + "grad_norm": 1.332981564422935, + "learning_rate": 1.0705383806982606e-08, + "loss": 0.521, + "step": 4122 + }, + { + "epoch": 0.9800332778702163, + "grad_norm": 1.4327146409945204, + "learning_rate": 1.0455061442548597e-08, + "loss": 0.5586, + "step": 4123 + }, + { + "epoch": 0.9802709769431899, + "grad_norm": 1.5142990160198728, + "learning_rate": 1.0207697365326452e-08, + "loss": 0.5912, + "step": 4124 + }, + { + "epoch": 0.9805086760161635, + "grad_norm": 1.322865638283676, + "learning_rate": 9.963291721978296e-09, + "loss": 0.5312, + "step": 4125 + }, + { + "epoch": 0.9807463750891372, + "grad_norm": 1.251415805392001, + "learning_rate": 9.72184465741155e-09, + "loss": 0.5382, + "step": 4126 + }, + { + "epoch": 0.9809840741621108, + "grad_norm": 1.317574830173402, + "learning_rate": 9.48335631477948e-09, + "loss": 0.5233, + "step": 4127 + }, + { + "epoch": 0.9812217732350844, + "grad_norm": 1.1384957209177864, + "learning_rate": 9.2478268354812e-09, + "loss": 0.5707, + "step": 4128 + }, + { + "epoch": 0.981459472308058, + "grad_norm": 1.3616007707456017, + "learning_rate": 9.015256359161118e-09, + "loss": 0.5251, + "step": 4129 + }, + { + "epoch": 0.9816971713810316, + "grad_norm": 1.2660635398392452, + "learning_rate": 8.785645023709489e-09, + "loss": 0.5993, + "step": 4130 + }, + { + "epoch": 0.9819348704540052, + "grad_norm": 1.3502266615028453, + "learning_rate": 8.558992965262969e-09, + "loss": 0.599, + "step": 4131 + }, + { + "epoch": 0.9821725695269788, + "grad_norm": 1.2856773849552838, + "learning_rate": 8.335300318201844e-09, + "loss": 0.5646, + "step": 4132 + }, + { + "epoch": 0.9824102685999525, + "grad_norm": 1.1017795144829032, + "learning_rate": 8.114567215154468e-09, + "loss": 0.5515, + "step": 4133 + }, + { + "epoch": 0.9826479676729261, + "grad_norm": 1.1743641720575446, + "learning_rate": 7.896793786991152e-09, + "loss": 0.5103, + "step": 4134 + }, + { + "epoch": 0.9828856667458997, + "grad_norm": 1.1384719993030143, + "learning_rate": 7.681980162830283e-09, + "loss": 0.54, + "step": 4135 + }, + { + "epoch": 0.9831233658188733, + "grad_norm": 1.1729582728245778, + "learning_rate": 7.470126470033868e-09, + "loss": 0.4895, + "step": 4136 + }, + { + "epoch": 0.9833610648918469, + "grad_norm": 1.2409482924312112, + "learning_rate": 7.261232834209208e-09, + "loss": 0.5907, + "step": 4137 + }, + { + "epoch": 0.9835987639648205, + "grad_norm": 1.2085026097116223, + "learning_rate": 7.055299379208902e-09, + "loss": 0.5374, + "step": 4138 + }, + { + "epoch": 0.9838364630377942, + "grad_norm": 1.2129814471817466, + "learning_rate": 6.852326227130835e-09, + "loss": 0.5755, + "step": 4139 + }, + { + "epoch": 0.9840741621107678, + "grad_norm": 1.309655341862937, + "learning_rate": 6.652313498316521e-09, + "loss": 0.5729, + "step": 4140 + }, + { + "epoch": 0.9843118611837414, + "grad_norm": 1.2449435750296454, + "learning_rate": 6.455261311352767e-09, + "loss": 0.5634, + "step": 4141 + }, + { + "epoch": 0.984549560256715, + "grad_norm": 1.6041793634138988, + "learning_rate": 6.2611697830722295e-09, + "loss": 0.5782, + "step": 4142 + }, + { + "epoch": 0.9847872593296886, + "grad_norm": 1.290135106781688, + "learning_rate": 6.070039028550634e-09, + "loss": 0.5077, + "step": 4143 + }, + { + "epoch": 0.9850249584026622, + "grad_norm": 1.2555843905211368, + "learning_rate": 5.881869161108444e-09, + "loss": 0.5789, + "step": 4144 + }, + { + "epoch": 0.9852626574756358, + "grad_norm": 1.2797589200846522, + "learning_rate": 5.696660292311418e-09, + "loss": 0.5587, + "step": 4145 + }, + { + "epoch": 0.9855003565486095, + "grad_norm": 1.2346839954464213, + "learning_rate": 5.514412531970048e-09, + "loss": 0.5587, + "step": 4146 + }, + { + "epoch": 0.9857380556215831, + "grad_norm": 1.2684453466751293, + "learning_rate": 5.3351259881379016e-09, + "loss": 0.5521, + "step": 4147 + }, + { + "epoch": 0.9859757546945567, + "grad_norm": 1.1286773693363559, + "learning_rate": 5.158800767113281e-09, + "loss": 0.5728, + "step": 4148 + }, + { + "epoch": 0.9862134537675303, + "grad_norm": 1.214957096434118, + "learning_rate": 4.985436973439228e-09, + "loss": 0.5885, + "step": 4149 + }, + { + "epoch": 0.9864511528405039, + "grad_norm": 1.2330630263183906, + "learning_rate": 4.815034709902411e-09, + "loss": 0.5214, + "step": 4150 + }, + { + "epoch": 0.9866888519134775, + "grad_norm": 1.1897445908418256, + "learning_rate": 4.647594077534235e-09, + "loss": 0.5214, + "step": 4151 + }, + { + "epoch": 0.9869265509864511, + "grad_norm": 1.427248786686241, + "learning_rate": 4.4831151756091766e-09, + "loss": 0.5225, + "step": 4152 + }, + { + "epoch": 0.9871642500594248, + "grad_norm": 1.3239578079891983, + "learning_rate": 4.321598101647007e-09, + "loss": 0.5523, + "step": 4153 + }, + { + "epoch": 0.9874019491323984, + "grad_norm": 1.3369297387332382, + "learning_rate": 4.163042951410567e-09, + "loss": 0.5525, + "step": 4154 + }, + { + "epoch": 0.987639648205372, + "grad_norm": 1.188056027982662, + "learning_rate": 4.007449818906328e-09, + "loss": 0.541, + "step": 4155 + }, + { + "epoch": 0.9878773472783456, + "grad_norm": 1.2580716721252414, + "learning_rate": 3.854818796385495e-09, + "loss": 0.5238, + "step": 4156 + }, + { + "epoch": 0.9881150463513192, + "grad_norm": 1.380293686382777, + "learning_rate": 3.705149974342348e-09, + "loss": 0.5493, + "step": 4157 + }, + { + "epoch": 0.9883527454242929, + "grad_norm": 1.3794648607626894, + "learning_rate": 3.558443441514792e-09, + "loss": 0.5985, + "step": 4158 + }, + { + "epoch": 0.9885904444972664, + "grad_norm": 1.1959337203286744, + "learning_rate": 3.41469928488547e-09, + "loss": 0.5269, + "step": 4159 + }, + { + "epoch": 0.9888281435702401, + "grad_norm": 1.3238948121309289, + "learning_rate": 3.273917589679543e-09, + "loss": 0.5619, + "step": 4160 + }, + { + "epoch": 0.9890658426432137, + "grad_norm": 1.4580738275114546, + "learning_rate": 3.1360984393657978e-09, + "loss": 0.5653, + "step": 4161 + }, + { + "epoch": 0.9893035417161873, + "grad_norm": 1.2936183373965529, + "learning_rate": 3.0012419156572047e-09, + "loss": 0.5414, + "step": 4162 + }, + { + "epoch": 0.989541240789161, + "grad_norm": 1.2629164297423225, + "learning_rate": 2.8693480985098056e-09, + "loss": 0.5461, + "step": 4163 + }, + { + "epoch": 0.9897789398621345, + "grad_norm": 1.1336567494272163, + "learning_rate": 2.7404170661232687e-09, + "loss": 0.5238, + "step": 4164 + }, + { + "epoch": 0.9900166389351082, + "grad_norm": 1.5247562077482941, + "learning_rate": 2.6144488949392253e-09, + "loss": 0.6223, + "step": 4165 + }, + { + "epoch": 0.9902543380080817, + "grad_norm": 1.3363109624590621, + "learning_rate": 2.491443659645709e-09, + "loss": 0.5821, + "step": 4166 + }, + { + "epoch": 0.9904920370810554, + "grad_norm": 1.1240992506258562, + "learning_rate": 2.371401433170495e-09, + "loss": 0.5308, + "step": 4167 + }, + { + "epoch": 0.990729736154029, + "grad_norm": 1.2666398498630282, + "learning_rate": 2.254322286687205e-09, + "loss": 0.5382, + "step": 4168 + }, + { + "epoch": 0.9909674352270026, + "grad_norm": 1.487872117841928, + "learning_rate": 2.140206289611979e-09, + "loss": 0.5465, + "step": 4169 + }, + { + "epoch": 0.9912051342999763, + "grad_norm": 1.287090069356833, + "learning_rate": 2.0290535096029184e-09, + "loss": 0.5319, + "step": 4170 + }, + { + "epoch": 0.9914428333729498, + "grad_norm": 1.2183112159685923, + "learning_rate": 1.9208640125628618e-09, + "loss": 0.5471, + "step": 4171 + }, + { + "epoch": 0.9916805324459235, + "grad_norm": 1.313641846356493, + "learning_rate": 1.815637862637165e-09, + "loss": 0.5589, + "step": 4172 + }, + { + "epoch": 0.9919182315188971, + "grad_norm": 1.1396837456591316, + "learning_rate": 1.7133751222137007e-09, + "loss": 0.5018, + "step": 4173 + }, + { + "epoch": 0.9921559305918707, + "grad_norm": 1.4275239399508812, + "learning_rate": 1.6140758519239685e-09, + "loss": 0.5553, + "step": 4174 + }, + { + "epoch": 0.9923936296648443, + "grad_norm": 1.3464562183475888, + "learning_rate": 1.5177401106419853e-09, + "loss": 0.4814, + "step": 4175 + }, + { + "epoch": 0.9926313287378179, + "grad_norm": 1.2907801205424583, + "learning_rate": 1.4243679554853952e-09, + "loss": 0.5709, + "step": 4176 + }, + { + "epoch": 0.9928690278107916, + "grad_norm": 1.1306721045166466, + "learning_rate": 1.3339594418138036e-09, + "loss": 0.5427, + "step": 4177 + }, + { + "epoch": 0.9931067268837651, + "grad_norm": 1.2714198743567573, + "learning_rate": 1.2465146232304436e-09, + "loss": 0.5318, + "step": 4178 + }, + { + "epoch": 0.9933444259567388, + "grad_norm": 1.9395942271101103, + "learning_rate": 1.1620335515810654e-09, + "loss": 0.5796, + "step": 4179 + }, + { + "epoch": 0.9935821250297124, + "grad_norm": 1.2161106283478613, + "learning_rate": 1.0805162769544908e-09, + "loss": 0.5031, + "step": 4180 + }, + { + "epoch": 0.993819824102686, + "grad_norm": 1.436155722540506, + "learning_rate": 1.0019628476815035e-09, + "loss": 0.5423, + "step": 4181 + }, + { + "epoch": 0.9940575231756597, + "grad_norm": 1.3161296844190553, + "learning_rate": 9.263733103365147e-10, + "loss": 0.5189, + "step": 4182 + }, + { + "epoch": 0.9942952222486332, + "grad_norm": 1.4330538538350912, + "learning_rate": 8.537477097364522e-10, + "loss": 0.5635, + "step": 4183 + }, + { + "epoch": 0.9945329213216069, + "grad_norm": 1.1955695239806272, + "learning_rate": 7.840860889407609e-10, + "loss": 0.5278, + "step": 4184 + }, + { + "epoch": 0.9947706203945804, + "grad_norm": 1.2444425562658372, + "learning_rate": 7.173884892519578e-10, + "loss": 0.4837, + "step": 4185 + }, + { + "epoch": 0.9950083194675541, + "grad_norm": 1.3127126473510289, + "learning_rate": 6.536549502145218e-10, + "loss": 0.5664, + "step": 4186 + }, + { + "epoch": 0.9952460185405277, + "grad_norm": 1.3485286367875644, + "learning_rate": 5.928855096154485e-10, + "loss": 0.5259, + "step": 4187 + }, + { + "epoch": 0.9954837176135013, + "grad_norm": 1.473721122454695, + "learning_rate": 5.350802034853608e-10, + "loss": 0.5553, + "step": 4188 + }, + { + "epoch": 0.995721416686475, + "grad_norm": 1.3796744750698187, + "learning_rate": 4.802390660968437e-10, + "loss": 0.5397, + "step": 4189 + }, + { + "epoch": 0.9959591157594485, + "grad_norm": 1.2119783030563933, + "learning_rate": 4.283621299649987e-10, + "loss": 0.5349, + "step": 4190 + }, + { + "epoch": 0.9961968148324222, + "grad_norm": 1.4166648125441106, + "learning_rate": 3.7944942584688947e-10, + "loss": 0.5863, + "step": 4191 + }, + { + "epoch": 0.9964345139053957, + "grad_norm": 1.1866268918448561, + "learning_rate": 3.335009827437619e-10, + "loss": 0.5077, + "step": 4192 + }, + { + "epoch": 0.9966722129783694, + "grad_norm": 1.2743943340190196, + "learning_rate": 2.905168278977133e-10, + "loss": 0.4979, + "step": 4193 + }, + { + "epoch": 0.996909912051343, + "grad_norm": 1.4594752739751926, + "learning_rate": 2.5049698679446843e-10, + "loss": 0.5523, + "step": 4194 + }, + { + "epoch": 0.9971476111243166, + "grad_norm": 1.4641885541794009, + "learning_rate": 2.1344148316060352e-10, + "loss": 0.5229, + "step": 4195 + }, + { + "epoch": 0.9973853101972903, + "grad_norm": 1.3644151755228935, + "learning_rate": 1.7935033896743225e-10, + "loss": 0.5465, + "step": 4196 + }, + { + "epoch": 0.9976230092702638, + "grad_norm": 1.1837296499854917, + "learning_rate": 1.4822357442656475e-10, + "loss": 0.5047, + "step": 4197 + }, + { + "epoch": 0.9978607083432375, + "grad_norm": 1.707836879593638, + "learning_rate": 1.200612079937935e-10, + "loss": 0.5373, + "step": 4198 + }, + { + "epoch": 0.998098407416211, + "grad_norm": 1.0560110512431837, + "learning_rate": 9.486325636576254e-11, + "loss": 0.5314, + "step": 4199 + }, + { + "epoch": 0.9983361064891847, + "grad_norm": 1.3471999649332567, + "learning_rate": 7.262973448218802e-11, + "loss": 0.5451, + "step": 4200 + }, + { + "epoch": 0.9983361064891847, + "eval_loss": 0.551862359046936, + "eval_runtime": 285.6656, + "eval_samples_per_second": 38.079, + "eval_steps_per_second": 1.19, + "step": 4200 + }, + { + "epoch": 0.9985738055621584, + "grad_norm": 1.3055978119406637, + "learning_rate": 5.336065552641323e-11, + "loss": 0.5664, + "step": 4201 + }, + { + "epoch": 0.9988115046351319, + "grad_norm": 1.6771147782672227, + "learning_rate": 3.7056030921522877e-11, + "loss": 0.5707, + "step": 4202 + }, + { + "epoch": 0.9990492037081056, + "grad_norm": 1.3112395552961387, + "learning_rate": 2.3715870335894175e-11, + "loss": 0.5587, + "step": 4203 + }, + { + "epoch": 0.9992869027810791, + "grad_norm": 1.1915239339135035, + "learning_rate": 1.3340181677645724e-11, + "loss": 0.5069, + "step": 4204 + }, + { + "epoch": 0.9995246018540528, + "grad_norm": 1.1855266020911146, + "learning_rate": 5.928971099078418e-12, + "loss": 0.5157, + "step": 4205 + }, + { + "epoch": 0.9997623009270263, + "grad_norm": 1.615654523502258, + "learning_rate": 1.4822429944549853e-12, + "loss": 0.5192, + "step": 4206 + }, + { + "epoch": 1.0, + "grad_norm": 1.2750803334801175, + "learning_rate": 0.0, + "loss": 0.5356, + "step": 4207 + }, + { + "epoch": 1.0, + "step": 4207, + "total_flos": 1.409292586450944e+16, + "train_loss": 0.23767077140749165, + "train_runtime": 54300.9785, + "train_samples_per_second": 19.834, + "train_steps_per_second": 0.077 + } + ], + "logging_steps": 1.0, + "max_steps": 4207, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 1.409292586450944e+16, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..604c0ba --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5ee1a0d450cb8ec873620be68a6eb25fde671a615b1e8e7e18e7bfa191458f9d +size 7096