初始化项目，由ModelHub XC社区提供模型
Model: karthik-2905/AL1-model-B Source: Original Platform
This commit is contained in:
ModelHub XC
48
.gitattributes
vendored
Normal file
48
.gitattributes
vendored
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
*.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
|
||||||
|
adapters/sft-lora/checkpoint-135/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sft-lora/checkpoint-45/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sft-lora/checkpoint-90/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sft-lora/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sft-lora-1ep/checkpoint-45/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sft-lora-1ep/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sweep-1e4/checkpoint-5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sweep-1e4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sweep-2e4/checkpoint-5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sweep-2e4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sweep-3e4/checkpoint-5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
adapters/sweep-3e4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
|
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||||
137
README.md
Normal file
137
README.md
Normal file
@@ -0,0 +1,137 @@
|
|||||||
|
---
|
||||||
|
license: apache-2.0
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: transformers
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
language:
|
||||||
|
- en
|
||||||
|
tags:
|
||||||
|
- tool-calling
|
||||||
|
- function-calling
|
||||||
|
- qwen3
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- trl
|
||||||
|
- zynthetix
|
||||||
|
---
|
||||||
|
|
||||||
|
# AL1-model-B — Qwen3-0.6B fine-tuned for native tool-calling
|
||||||
|
|
||||||
|
A compact (0.6B-parameter) language model fine-tuned to **call tools reliably** while staying a
|
||||||
|
capable general-chat assistant. It takes the Apache-2.0 [`Qwen/Qwen3-0.6B`](https://huggingface.co/Qwen/Qwen3-0.6B)
|
||||||
|
base and applies a single **LoRA supervised fine-tune (SFT)** targeting native, Hermes-style
|
||||||
|
`<tool_call>` emission over a small set of file/shell tools — then merges the adapter back into
|
||||||
|
the base for a self-contained model.
|
||||||
|
|
||||||
|
Part of the **Zynthetix AL-1** two-model lab (Model A = from-scratch NumPy LM; **Model B = improve an
|
||||||
|
open-source small model**, this repo). Code: https://github.com/GruheshKurra/AL1-model-B
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Results — before → after
|
||||||
|
|
||||||
|
Tool-calling is scored by **exact match** against a held-out 12-case set: `name_acc` = correct tool
|
||||||
|
name chosen, `full_acc` = correct name **and** arguments. Both evaluated with greedy decoding, fp32,
|
||||||
|
on the same device (no blending with chat scores).
|
||||||
|
|
||||||
|
| Metric | Base `Qwen3-0.6B` | **AL1-model-B (SFT)** | Δ |
|
||||||
|
|---|---|---|---|
|
||||||
|
| Tool `name_acc` | 0.500 | **1.000** | +0.500 |
|
||||||
|
| Tool `full_acc` | 0.500 | **0.917** (11/12) | +0.417 |
|
||||||
|
| Chat coherence (10 probes) | strong | strong, terser | no regression |
|
||||||
|
|
||||||
|
**Why the base scored only 0.5:** it *refused* to call `run_bash` and `edit_file` (0/2 each) and
|
||||||
|
mis-handled one `list_dir` — its arguments were fine *when* it chose to call. The SFT data therefore
|
||||||
|
over-weighted `edit_file` + `run_bash` + explicit anti-refusal examples, which closed the entire gap
|
||||||
|
(`name_acc` → 1.0 = a valid tool is now always selected).
|
||||||
|
|
||||||
|
> The merged fp16 model reproduces the adapter score exactly (`full_acc` 0.917) — merging lost nothing.
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## The tools
|
||||||
|
|
||||||
|
Five file/shell tools, described to the model via the OpenAI-style `tools=` schema and the Qwen3
|
||||||
|
native chat template (Hermes `<tool_call>` tags):
|
||||||
|
|
||||||
|
| Tool | Purpose | Required args |
|
||||||
|
|---|---|---|
|
||||||
|
| `read_file` | Read a file's contents | `path` |
|
||||||
|
| `write_file` | Create/overwrite a file | `path`, `content` |
|
||||||
|
| `edit_file` | Replace an exact substring | `path`, `old_string`, `new_string` |
|
||||||
|
| `list_dir` | List a directory | `path` |
|
||||||
|
| `run_bash` | Run a shell command | `command` |
|
||||||
|
|
||||||
|
The model **emits** tool name + arguments; wiring those to a real executor (or an MCP client) is a
|
||||||
|
serving-layer concern, deliberately out of scope for the model itself.
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Usage
|
||||||
|
|
||||||
|
```python
|
||||||
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||||
|
import torch
|
||||||
|
|
||||||
|
MODEL = "karthik-2905/AL1-model-B"
|
||||||
|
tok = AutoTokenizer.from_pretrained(MODEL)
|
||||||
|
model = AutoModelForCausalLM.from_pretrained(MODEL, dtype=torch.float16).to("mps") # or "cuda"/"cpu"
|
||||||
|
|
||||||
|
TOOLS = [{"type": "function", "function": {
|
||||||
|
"name": "read_file",
|
||||||
|
"description": "Read and return the contents of a file.",
|
||||||
|
"parameters": {"type": "object",
|
||||||
|
"properties": {"path": {"type": "string"}}, "required": ["path"]}}}] # + the other 4
|
||||||
|
|
||||||
|
msgs = [{"role": "user", "content": "Show me what's in config.py"}]
|
||||||
|
text = tok.apply_chat_template(msgs, tools=TOOLS, tokenize=False,
|
||||||
|
add_generation_prompt=True, enable_thinking=False)
|
||||||
|
ids = tok(text, return_tensors="pt").to(model.device)
|
||||||
|
out = model.generate(**ids, max_new_tokens=128, do_sample=False)
|
||||||
|
print(tok.decode(out[0][ids.input_ids.shape[1]:], skip_special_tokens=True))
|
||||||
|
# -> <tool_call>{"name": "read_file", "arguments": {"path": "config.py"}}</tool_call>
|
||||||
|
```
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## How it was built
|
||||||
|
|
||||||
|
| Stage | Choice |
|
||||||
|
|---|---|
|
||||||
|
| Base | `Qwen/Qwen3-0.6B` (Apache-2.0) — GQA + RoPE + SwiGLU + RMSNorm |
|
||||||
|
| PEFT | LoRA `r=16`, `α=32`, dropout 0.05, on all 7 attention+MLP projections |
|
||||||
|
| Trainer | TRL `SFTTrainer`, `assistant_only_loss=True`, `max_length=512`, packing off |
|
||||||
|
| Schedule | lr `3e-4` (cosine, 3% warmup), 3 epochs, effective batch 32, bf16 |
|
||||||
|
| Hardware | RunPod **RTX A6000** (train), Apple Silicon / MPS (merge + eval) |
|
||||||
|
| Merge | `merge_and_unload()` → standalone fp16 |
|
||||||
|
|
||||||
|
**LR sweep** (10% data, 1 epoch) picked `3e-4` (eval_loss: `1e-4`=1.099, `2e-4`=0.999, `3e-4`=0.975).
|
||||||
|
The full 3-epoch run overfits by *eval_loss* after ~epoch 1 (0.91 → 0.99), but generation quality and
|
||||||
|
tool exact-match were best at 3 epochs — the **task metric beat the proxy loss**, so 3 epochs shipped.
|
||||||
|
|
||||||
|
### Data
|
||||||
|
|
||||||
|
- **1,423 train / 74 val** examples, chat-formatted to the Qwen3 template, ~**2:1 tool:chat** ratio.
|
||||||
|
- Tool examples teach the Hermes `<tool_call>` format across all 5 tools (over-weighting the two the
|
||||||
|
base refused); chat examples guard against chat regression.
|
||||||
|
- Train/val split is dedup'd and **leakage-checked against the eval set**.
|
||||||
|
|
||||||
|
### Evaluation
|
||||||
|
|
||||||
|
- **Tool track (trust anchor):** exact-match over a 12-case set, scored by `mlb/eval_tools.py`.
|
||||||
|
- **Chat track (regression guard):** 10 probes, currently eyeballed for coherence (LLM-judge = future work).
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Limitations
|
||||||
|
|
||||||
|
- **Small model (0.6B):** capable within its size, not a substitute for a large instruct model.
|
||||||
|
- **Chat quality is eyeballed,** not yet LLM-judge scored — no formal regression number.
|
||||||
|
- **No general-capability guard** (MMLU/GSM8K) was run; broad-knowledge regression is unmeasured.
|
||||||
|
- **4-bit quantization untested** — a 0.6B model can degrade more than the usual −1–2% at 4-bit.
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## License
|
||||||
|
|
||||||
|
**Apache-2.0**, inherited from the `Qwen/Qwen3-0.6B` base.
|
||||||
62
adapters/sft-lora-1ep/README.md
Normal file
62
adapters/sft-lora-1ep/README.md
Normal file
@@ -0,0 +1,62 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
model_name: sft-lora-1ep
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
licence: license
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for sft-lora-1ep
|
||||||
|
|
||||||
|
This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B).
|
||||||
|
It has been trained using [TRL](https://github.com/huggingface/trl).
|
||||||
|
|
||||||
|
## Quick start
|
||||||
|
|
||||||
|
```python
|
||||||
|
from transformers import pipeline
|
||||||
|
|
||||||
|
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
||||||
|
generator = pipeline("text-generation", model="None", device="cuda")
|
||||||
|
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
||||||
|
print(output["generated_text"])
|
||||||
|
```
|
||||||
|
|
||||||
|
## Training procedure
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
This model was trained with SFT.
|
||||||
|
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
|
- TRL: 1.8.0
|
||||||
|
- Transformers: 5.13.1
|
||||||
|
- Pytorch: 2.8.0+cu128
|
||||||
|
- Datasets: 5.0.0
|
||||||
|
- Tokenizers: 0.22.2
|
||||||
|
|
||||||
|
## Citations
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
Cite TRL as:
|
||||||
|
|
||||||
|
```bibtex
|
||||||
|
@software{vonwerra2020trl,
|
||||||
|
title = {{TRL: Transformers Reinforcement Learning}},
|
||||||
|
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
||||||
|
license = {Apache-2.0},
|
||||||
|
url = {https://github.com/huggingface/trl},
|
||||||
|
year = {2020}
|
||||||
|
}
|
||||||
|
```
|
||||||
48
adapters/sft-lora-1ep/adapter_config.json
Normal file
48
adapters/sft-lora-1ep/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"o_proj",
|
||||||
|
"down_proj",
|
||||||
|
"v_proj",
|
||||||
|
"gate_proj",
|
||||||
|
"q_proj",
|
||||||
|
"k_proj",
|
||||||
|
"up_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
3
adapters/sft-lora-1ep/adapter_model.safetensors
Normal file
3
adapters/sft-lora-1ep/adapter_model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:47c379f9a9f9eda2d1d4c3194060ab471ea74f9548edd049a500127bb93f7bc6
|
||||||
|
size 40422168
|
||||||
89
adapters/sft-lora-1ep/chat_template.jinja
Normal file
89
adapters/sft-lora-1ep/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
209
adapters/sft-lora-1ep/checkpoint-45/README.md
Normal file
209
adapters/sft-lora-1ep/checkpoint-45/README.md
Normal file
@@ -0,0 +1,209 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for Model ID
|
||||||
|
|
||||||
|
<!-- Provide a quick summary of what the model is/does. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Details
|
||||||
|
|
||||||
|
### Model Description
|
||||||
|
|
||||||
|
<!-- Provide a longer summary of what this model is. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
- **Developed by:** [More Information Needed]
|
||||||
|
- **Funded by [optional]:** [More Information Needed]
|
||||||
|
- **Shared by [optional]:** [More Information Needed]
|
||||||
|
- **Model type:** [More Information Needed]
|
||||||
|
- **Language(s) (NLP):** [More Information Needed]
|
||||||
|
- **License:** [More Information Needed]
|
||||||
|
- **Finetuned from model [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
### Model Sources [optional]
|
||||||
|
|
||||||
|
<!-- Provide the basic links for the model. -->
|
||||||
|
|
||||||
|
- **Repository:** [More Information Needed]
|
||||||
|
- **Paper [optional]:** [More Information Needed]
|
||||||
|
- **Demo [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
## Uses
|
||||||
|
|
||||||
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
||||||
|
|
||||||
|
### Direct Use
|
||||||
|
|
||||||
|
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Downstream Use [optional]
|
||||||
|
|
||||||
|
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Out-of-Scope Use
|
||||||
|
|
||||||
|
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Bias, Risks, and Limitations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Recommendations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
||||||
|
|
||||||
|
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
||||||
|
|
||||||
|
## How to Get Started with the Model
|
||||||
|
|
||||||
|
Use the code below to get started with the model.
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Training Details
|
||||||
|
|
||||||
|
### Training Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Training Procedure
|
||||||
|
|
||||||
|
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
||||||
|
|
||||||
|
#### Preprocessing [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
|
||||||
|
#### Training Hyperparameters
|
||||||
|
|
||||||
|
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
||||||
|
|
||||||
|
#### Speeds, Sizes, Times [optional]
|
||||||
|
|
||||||
|
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Evaluation
|
||||||
|
|
||||||
|
<!-- This section describes the evaluation protocols and provides the results. -->
|
||||||
|
|
||||||
|
### Testing Data, Factors & Metrics
|
||||||
|
|
||||||
|
#### Testing Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card if possible. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Factors
|
||||||
|
|
||||||
|
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Metrics
|
||||||
|
|
||||||
|
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Results
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Summary
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Examination [optional]
|
||||||
|
|
||||||
|
<!-- Relevant interpretability work for the model goes here -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Environmental Impact
|
||||||
|
|
||||||
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
||||||
|
|
||||||
|
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
||||||
|
|
||||||
|
- **Hardware Type:** [More Information Needed]
|
||||||
|
- **Hours used:** [More Information Needed]
|
||||||
|
- **Cloud Provider:** [More Information Needed]
|
||||||
|
- **Compute Region:** [More Information Needed]
|
||||||
|
- **Carbon Emitted:** [More Information Needed]
|
||||||
|
|
||||||
|
## Technical Specifications [optional]
|
||||||
|
|
||||||
|
### Model Architecture and Objective
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Compute Infrastructure
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Hardware
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Software
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Citation [optional]
|
||||||
|
|
||||||
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
||||||
|
|
||||||
|
**BibTeX:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
**APA:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Glossary [optional]
|
||||||
|
|
||||||
|
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## More Information [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Authors [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Contact
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
48
adapters/sft-lora-1ep/checkpoint-45/adapter_config.json
Normal file
48
adapters/sft-lora-1ep/checkpoint-45/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"o_proj",
|
||||||
|
"down_proj",
|
||||||
|
"v_proj",
|
||||||
|
"gate_proj",
|
||||||
|
"q_proj",
|
||||||
|
"k_proj",
|
||||||
|
"up_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:47c379f9a9f9eda2d1d4c3194060ab471ea74f9548edd049a500127bb93f7bc6
|
||||||
|
size 40422168
|
||||||
89
adapters/sft-lora-1ep/checkpoint-45/chat_template.jinja
Normal file
89
adapters/sft-lora-1ep/checkpoint-45/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
3
adapters/sft-lora-1ep/checkpoint-45/optimizer.pt
Normal file
3
adapters/sft-lora-1ep/checkpoint-45/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:566faa9c7d439e880d09e4019aa0f41acc4ecdc2cdb368a5f1b465ab84aa66e8
|
||||||
|
size 81075515
|
||||||
3
adapters/sft-lora-1ep/checkpoint-45/rng_state.pth
Normal file
3
adapters/sft-lora-1ep/checkpoint-45/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:8afaaaf68ec369c33e52739656f45b852ca8ff769f8b378c0e061a55f3a6dd8d
|
||||||
|
size 14645
|
||||||
3
adapters/sft-lora-1ep/checkpoint-45/scheduler.pt
Normal file
3
adapters/sft-lora-1ep/checkpoint-45/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:ba53d412fdaef3ea9e83daaefc9058ccddd307d410e5fc8901e13fadf373c25b
|
||||||
|
size 1465
|
||||||
3
adapters/sft-lora-1ep/checkpoint-45/tokenizer.json
Normal file
3
adapters/sft-lora-1ep/checkpoint-45/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sft-lora-1ep/checkpoint-45/tokenizer_config.json
Normal file
30
adapters/sft-lora-1ep/checkpoint-45/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
179
adapters/sft-lora-1ep/checkpoint-45/trainer_state.json
Normal file
179
adapters/sft-lora-1ep/checkpoint-45/trainer_state.json
Normal file
@@ -0,0 +1,179 @@
|
|||||||
|
{
|
||||||
|
"best_global_step": null,
|
||||||
|
"best_metric": null,
|
||||||
|
"best_model_checkpoint": null,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"eval_steps": 10,
|
||||||
|
"global_step": 45,
|
||||||
|
"is_hyper_param_search": false,
|
||||||
|
"is_local_process_zero": true,
|
||||||
|
"is_world_process_zero": true,
|
||||||
|
"log_history": [
|
||||||
|
{
|
||||||
|
"entropy": 0.5051268041133881,
|
||||||
|
"epoch": 0.1111111111111111,
|
||||||
|
"grad_norm": 1.7442580461502075,
|
||||||
|
"learning_rate": 0.0002984015052112665,
|
||||||
|
"loss": 1.0908665657043457,
|
||||||
|
"mean_token_accuracy": 0.8142835974693299,
|
||||||
|
"num_tokens": 59002.0,
|
||||||
|
"step": 5
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7225139260292053,
|
||||||
|
"epoch": 0.2222222222222222,
|
||||||
|
"grad_norm": 0.7206196784973145,
|
||||||
|
"learning_rate": 0.0002808074112215711,
|
||||||
|
"loss": 0.6919548034667968,
|
||||||
|
"mean_token_accuracy": 0.8514000058174134,
|
||||||
|
"num_tokens": 120432.0,
|
||||||
|
"step": 10
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.2222222222222222,
|
||||||
|
"eval_entropy": 0.9578993916511536,
|
||||||
|
"eval_loss": 0.9403628706932068,
|
||||||
|
"eval_mean_token_accuracy": 0.8023552894592285,
|
||||||
|
"eval_num_tokens": 120432.0,
|
||||||
|
"eval_runtime": 1.6133,
|
||||||
|
"eval_samples_per_second": 45.869,
|
||||||
|
"eval_steps_per_second": 1.86,
|
||||||
|
"step": 10
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8998202800750732,
|
||||||
|
"epoch": 0.3333333333333333,
|
||||||
|
"grad_norm": 0.6158784031867981,
|
||||||
|
"learning_rate": 0.00024595095323383365,
|
||||||
|
"loss": 0.8696078300476074,
|
||||||
|
"mean_token_accuracy": 0.8144582986831665,
|
||||||
|
"num_tokens": 179678.0,
|
||||||
|
"step": 15
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7956753373146057,
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"grad_norm": 0.638497531414032,
|
||||||
|
"learning_rate": 0.00019843206071571692,
|
||||||
|
"loss": 0.8670864105224609,
|
||||||
|
"mean_token_accuracy": 0.8208897352218628,
|
||||||
|
"num_tokens": 240600.0,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"eval_entropy": 0.8266730904579163,
|
||||||
|
"eval_loss": 0.9242745637893677,
|
||||||
|
"eval_mean_token_accuracy": 0.803473949432373,
|
||||||
|
"eval_num_tokens": 240600.0,
|
||||||
|
"eval_runtime": 1.6168,
|
||||||
|
"eval_samples_per_second": 45.77,
|
||||||
|
"eval_steps_per_second": 1.856,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7963908970355987,
|
||||||
|
"epoch": 0.5555555555555556,
|
||||||
|
"grad_norm": 0.42324671149253845,
|
||||||
|
"learning_rate": 0.00014452169654135115,
|
||||||
|
"loss": 0.8695890426635742,
|
||||||
|
"mean_token_accuracy": 0.8264959454536438,
|
||||||
|
"num_tokens": 301638.0,
|
||||||
|
"step": 25
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.9203879475593567,
|
||||||
|
"epoch": 0.6666666666666666,
|
||||||
|
"grad_norm": 0.6087479591369629,
|
||||||
|
"learning_rate": 9.133429192647661e-05,
|
||||||
|
"loss": 0.9359363555908203,
|
||||||
|
"mean_token_accuracy": 0.8043835282325744,
|
||||||
|
"num_tokens": 359243.0,
|
||||||
|
"step": 30
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.6666666666666666,
|
||||||
|
"eval_entropy": 0.861892561117808,
|
||||||
|
"eval_loss": 0.918761670589447,
|
||||||
|
"eval_mean_token_accuracy": 0.8063527544339498,
|
||||||
|
"eval_num_tokens": 359243.0,
|
||||||
|
"eval_runtime": 1.6289,
|
||||||
|
"eval_samples_per_second": 45.43,
|
||||||
|
"eval_steps_per_second": 1.842,
|
||||||
|
"step": 30
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6894854545593262,
|
||||||
|
"epoch": 0.7777777777777778,
|
||||||
|
"grad_norm": 0.48022472858428955,
|
||||||
|
"learning_rate": 4.58888707169049e-05,
|
||||||
|
"loss": 0.6996908187866211,
|
||||||
|
"mean_token_accuracy": 0.849899160861969,
|
||||||
|
"num_tokens": 417098.0,
|
||||||
|
"step": 35
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8869106888771057,
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"grad_norm": 0.5962172746658325,
|
||||||
|
"learning_rate": 1.4182764376028006e-05,
|
||||||
|
"loss": 0.9062419891357422,
|
||||||
|
"mean_token_accuracy": 0.8095215201377869,
|
||||||
|
"num_tokens": 480008.0,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"eval_entropy": 0.8640047709147135,
|
||||||
|
"eval_loss": 0.9166465997695923,
|
||||||
|
"eval_mean_token_accuracy": 0.8074178695678711,
|
||||||
|
"eval_num_tokens": 480008.0,
|
||||||
|
"eval_runtime": 1.6231,
|
||||||
|
"eval_samples_per_second": 45.593,
|
||||||
|
"eval_steps_per_second": 1.848,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8551887631416321,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"grad_norm": 0.8982498049736023,
|
||||||
|
"learning_rate": 4.0015745046725336e-07,
|
||||||
|
"loss": 0.8875165939331054,
|
||||||
|
"mean_token_accuracy": 0.8184282779693604,
|
||||||
|
"num_tokens": 530794.0,
|
||||||
|
"step": 45
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.0,
|
||||||
|
"eval_entropy": 0.8636427720387777,
|
||||||
|
"eval_loss": 0.9162904620170593,
|
||||||
|
"eval_mean_token_accuracy": 0.8047811388969421,
|
||||||
|
"eval_num_tokens": 530794.0,
|
||||||
|
"eval_runtime": 1.6244,
|
||||||
|
"eval_samples_per_second": 45.555,
|
||||||
|
"eval_steps_per_second": 1.847,
|
||||||
|
"step": 45
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"logging_steps": 5,
|
||||||
|
"max_steps": 45,
|
||||||
|
"num_input_tokens_seen": 0,
|
||||||
|
"num_train_epochs": 1,
|
||||||
|
"save_steps": 500,
|
||||||
|
"stateful_callbacks": {
|
||||||
|
"TrainerControl": {
|
||||||
|
"args": {
|
||||||
|
"should_epoch_stop": false,
|
||||||
|
"should_evaluate": false,
|
||||||
|
"should_log": false,
|
||||||
|
"should_save": true,
|
||||||
|
"should_training_stop": true
|
||||||
|
},
|
||||||
|
"attributes": {}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"total_flos": 1952299008000000.0,
|
||||||
|
"train_batch_size": 32,
|
||||||
|
"trial_name": null,
|
||||||
|
"trial_params": null
|
||||||
|
}
|
||||||
3
adapters/sft-lora-1ep/checkpoint-45/training_args.bin
Normal file
3
adapters/sft-lora-1ep/checkpoint-45/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:924d7e63473ec418c24607d3b642e15bbb9db040609b99cad5c74bcf3d799620
|
||||||
|
size 5777
|
||||||
3
adapters/sft-lora-1ep/tokenizer.json
Normal file
3
adapters/sft-lora-1ep/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sft-lora-1ep/tokenizer_config.json
Normal file
30
adapters/sft-lora-1ep/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
3
adapters/sft-lora-1ep/training_args.bin
Normal file
3
adapters/sft-lora-1ep/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:924d7e63473ec418c24607d3b642e15bbb9db040609b99cad5c74bcf3d799620
|
||||||
|
size 5777
|
||||||
62
adapters/sft-lora/README.md
Normal file
62
adapters/sft-lora/README.md
Normal file
@@ -0,0 +1,62 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
model_name: sft-lora
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
licence: license
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for sft-lora
|
||||||
|
|
||||||
|
This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B).
|
||||||
|
It has been trained using [TRL](https://github.com/huggingface/trl).
|
||||||
|
|
||||||
|
## Quick start
|
||||||
|
|
||||||
|
```python
|
||||||
|
from transformers import pipeline
|
||||||
|
|
||||||
|
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
||||||
|
generator = pipeline("text-generation", model="None", device="cuda")
|
||||||
|
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
||||||
|
print(output["generated_text"])
|
||||||
|
```
|
||||||
|
|
||||||
|
## Training procedure
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
This model was trained with SFT.
|
||||||
|
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
|
- TRL: 1.8.0
|
||||||
|
- Transformers: 5.13.1
|
||||||
|
- Pytorch: 2.8.0+cu128
|
||||||
|
- Datasets: 5.0.0
|
||||||
|
- Tokenizers: 0.22.2
|
||||||
|
|
||||||
|
## Citations
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
Cite TRL as:
|
||||||
|
|
||||||
|
```bibtex
|
||||||
|
@software{vonwerra2020trl,
|
||||||
|
title = {{TRL: Transformers Reinforcement Learning}},
|
||||||
|
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
||||||
|
license = {Apache-2.0},
|
||||||
|
url = {https://github.com/huggingface/trl},
|
||||||
|
year = {2020}
|
||||||
|
}
|
||||||
|
```
|
||||||
48
adapters/sft-lora/adapter_config.json
Normal file
48
adapters/sft-lora/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"gate_proj",
|
||||||
|
"down_proj",
|
||||||
|
"o_proj",
|
||||||
|
"v_proj",
|
||||||
|
"k_proj",
|
||||||
|
"up_proj",
|
||||||
|
"q_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
3
adapters/sft-lora/adapter_model.safetensors
Normal file
3
adapters/sft-lora/adapter_model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:9f847ef4b95bbb96bb1d6a9b59dbcc914b2d630e9021bad6dc88bcc0c846cadc
|
||||||
|
size 40422168
|
||||||
89
adapters/sft-lora/chat_template.jinja
Normal file
89
adapters/sft-lora/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
209
adapters/sft-lora/checkpoint-135/README.md
Normal file
209
adapters/sft-lora/checkpoint-135/README.md
Normal file
@@ -0,0 +1,209 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for Model ID
|
||||||
|
|
||||||
|
<!-- Provide a quick summary of what the model is/does. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Details
|
||||||
|
|
||||||
|
### Model Description
|
||||||
|
|
||||||
|
<!-- Provide a longer summary of what this model is. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
- **Developed by:** [More Information Needed]
|
||||||
|
- **Funded by [optional]:** [More Information Needed]
|
||||||
|
- **Shared by [optional]:** [More Information Needed]
|
||||||
|
- **Model type:** [More Information Needed]
|
||||||
|
- **Language(s) (NLP):** [More Information Needed]
|
||||||
|
- **License:** [More Information Needed]
|
||||||
|
- **Finetuned from model [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
### Model Sources [optional]
|
||||||
|
|
||||||
|
<!-- Provide the basic links for the model. -->
|
||||||
|
|
||||||
|
- **Repository:** [More Information Needed]
|
||||||
|
- **Paper [optional]:** [More Information Needed]
|
||||||
|
- **Demo [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
## Uses
|
||||||
|
|
||||||
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
||||||
|
|
||||||
|
### Direct Use
|
||||||
|
|
||||||
|
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Downstream Use [optional]
|
||||||
|
|
||||||
|
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Out-of-Scope Use
|
||||||
|
|
||||||
|
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Bias, Risks, and Limitations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Recommendations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
||||||
|
|
||||||
|
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
||||||
|
|
||||||
|
## How to Get Started with the Model
|
||||||
|
|
||||||
|
Use the code below to get started with the model.
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Training Details
|
||||||
|
|
||||||
|
### Training Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Training Procedure
|
||||||
|
|
||||||
|
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
||||||
|
|
||||||
|
#### Preprocessing [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
|
||||||
|
#### Training Hyperparameters
|
||||||
|
|
||||||
|
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
||||||
|
|
||||||
|
#### Speeds, Sizes, Times [optional]
|
||||||
|
|
||||||
|
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Evaluation
|
||||||
|
|
||||||
|
<!-- This section describes the evaluation protocols and provides the results. -->
|
||||||
|
|
||||||
|
### Testing Data, Factors & Metrics
|
||||||
|
|
||||||
|
#### Testing Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card if possible. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Factors
|
||||||
|
|
||||||
|
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Metrics
|
||||||
|
|
||||||
|
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Results
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Summary
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Examination [optional]
|
||||||
|
|
||||||
|
<!-- Relevant interpretability work for the model goes here -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Environmental Impact
|
||||||
|
|
||||||
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
||||||
|
|
||||||
|
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
||||||
|
|
||||||
|
- **Hardware Type:** [More Information Needed]
|
||||||
|
- **Hours used:** [More Information Needed]
|
||||||
|
- **Cloud Provider:** [More Information Needed]
|
||||||
|
- **Compute Region:** [More Information Needed]
|
||||||
|
- **Carbon Emitted:** [More Information Needed]
|
||||||
|
|
||||||
|
## Technical Specifications [optional]
|
||||||
|
|
||||||
|
### Model Architecture and Objective
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Compute Infrastructure
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Hardware
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Software
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Citation [optional]
|
||||||
|
|
||||||
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
||||||
|
|
||||||
|
**BibTeX:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
**APA:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Glossary [optional]
|
||||||
|
|
||||||
|
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## More Information [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Authors [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Contact
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
48
adapters/sft-lora/checkpoint-135/adapter_config.json
Normal file
48
adapters/sft-lora/checkpoint-135/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"gate_proj",
|
||||||
|
"down_proj",
|
||||||
|
"o_proj",
|
||||||
|
"v_proj",
|
||||||
|
"k_proj",
|
||||||
|
"up_proj",
|
||||||
|
"q_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:9f847ef4b95bbb96bb1d6a9b59dbcc914b2d630e9021bad6dc88bcc0c846cadc
|
||||||
|
size 40422168
|
||||||
89
adapters/sft-lora/checkpoint-135/chat_template.jinja
Normal file
89
adapters/sft-lora/checkpoint-135/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
3
adapters/sft-lora/checkpoint-135/optimizer.pt
Normal file
3
adapters/sft-lora/checkpoint-135/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:14982afa7d229f65e42259eb79178fc662512463c82d9ee0560218c19e076081
|
||||||
|
size 81075515
|
||||||
3
adapters/sft-lora/checkpoint-135/rng_state.pth
Normal file
3
adapters/sft-lora/checkpoint-135/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:d40f49a4e26bcf9a43857449b0d4c7574c032d6aac208b6b90c99708b30cf0ee
|
||||||
|
size 14645
|
||||||
3
adapters/sft-lora/checkpoint-135/scheduler.pt
Normal file
3
adapters/sft-lora/checkpoint-135/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:0b6d991e21e78bd71614d8c7c4f42d8c2dedc292b93cac29c43d3e47ffc68474
|
||||||
|
size 1465
|
||||||
3
adapters/sft-lora/checkpoint-135/tokenizer.json
Normal file
3
adapters/sft-lora/checkpoint-135/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sft-lora/checkpoint-135/tokenizer_config.json
Normal file
30
adapters/sft-lora/checkpoint-135/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
381
adapters/sft-lora/checkpoint-135/trainer_state.json
Normal file
381
adapters/sft-lora/checkpoint-135/trainer_state.json
Normal file
@@ -0,0 +1,381 @@
|
|||||||
|
{
|
||||||
|
"best_global_step": null,
|
||||||
|
"best_metric": null,
|
||||||
|
"best_model_checkpoint": null,
|
||||||
|
"epoch": 3.0,
|
||||||
|
"eval_steps": 20,
|
||||||
|
"global_step": 135,
|
||||||
|
"is_hyper_param_search": false,
|
||||||
|
"is_local_process_zero": true,
|
||||||
|
"is_world_process_zero": true,
|
||||||
|
"log_history": [
|
||||||
|
{
|
||||||
|
"entropy": 0.46982616782188413,
|
||||||
|
"epoch": 0.1111111111111111,
|
||||||
|
"grad_norm": 1.9415351152420044,
|
||||||
|
"learning_rate": 0.00023999999999999998,
|
||||||
|
"loss": 1.166135025024414,
|
||||||
|
"mean_token_accuracy": 0.8076520800590515,
|
||||||
|
"num_tokens": 59002.0,
|
||||||
|
"step": 5
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.681049507856369,
|
||||||
|
"epoch": 0.2222222222222222,
|
||||||
|
"grad_norm": 0.8071348071098328,
|
||||||
|
"learning_rate": 0.00029929974520764727,
|
||||||
|
"loss": 0.7030046463012696,
|
||||||
|
"mean_token_accuracy": 0.849864411354065,
|
||||||
|
"num_tokens": 120432.0,
|
||||||
|
"step": 10
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.9322404026985168,
|
||||||
|
"epoch": 0.3333333333333333,
|
||||||
|
"grad_norm": 0.6664227247238159,
|
||||||
|
"learning_rate": 0.0002964661618243908,
|
||||||
|
"loss": 0.878734016418457,
|
||||||
|
"mean_token_accuracy": 0.8135401248931885,
|
||||||
|
"num_tokens": 179678.0,
|
||||||
|
"step": 15
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8074370503425599,
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"grad_norm": 0.6451402306556702,
|
||||||
|
"learning_rate": 0.0002914967719886614,
|
||||||
|
"loss": 0.8701633453369141,
|
||||||
|
"mean_token_accuracy": 0.8204897165298461,
|
||||||
|
"num_tokens": 240600.0,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"eval_entropy": 0.8143169482549032,
|
||||||
|
"eval_loss": 0.924848735332489,
|
||||||
|
"eval_mean_token_accuracy": 0.803371528784434,
|
||||||
|
"eval_num_tokens": 240600.0,
|
||||||
|
"eval_runtime": 1.6201,
|
||||||
|
"eval_samples_per_second": 45.676,
|
||||||
|
"eval_steps_per_second": 1.852,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7912694752216339,
|
||||||
|
"epoch": 0.5555555555555556,
|
||||||
|
"grad_norm": 0.4144524037837982,
|
||||||
|
"learning_rate": 0.0002844640405943555,
|
||||||
|
"loss": 0.8696941375732422,
|
||||||
|
"mean_token_accuracy": 0.8263819813728333,
|
||||||
|
"num_tokens": 301638.0,
|
||||||
|
"step": 25
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.9237359523773193,
|
||||||
|
"epoch": 0.6666666666666666,
|
||||||
|
"grad_norm": 0.5862515568733215,
|
||||||
|
"learning_rate": 0.00027547052070153396,
|
||||||
|
"loss": 0.9361928939819336,
|
||||||
|
"mean_token_accuracy": 0.8028241634368897,
|
||||||
|
"num_tokens": 359243.0,
|
||||||
|
"step": 30
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6837044417858124,
|
||||||
|
"epoch": 0.7777777777777778,
|
||||||
|
"grad_norm": 0.4594927728176117,
|
||||||
|
"learning_rate": 0.0002646473580818772,
|
||||||
|
"loss": 0.6992801666259766,
|
||||||
|
"mean_token_accuracy": 0.8513208270072937,
|
||||||
|
"num_tokens": 417098.0,
|
||||||
|
"step": 35
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8795680403709412,
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"grad_norm": 0.5710228681564331,
|
||||||
|
"learning_rate": 0.00025215237881801955,
|
||||||
|
"loss": 0.9085055351257324,
|
||||||
|
"mean_token_accuracy": 0.8098101854324341,
|
||||||
|
"num_tokens": 480008.0,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"eval_entropy": 0.8618650635083517,
|
||||||
|
"eval_loss": 0.9124361872673035,
|
||||||
|
"eval_mean_token_accuracy": 0.8028692404429117,
|
||||||
|
"eval_num_tokens": 480008.0,
|
||||||
|
"eval_runtime": 1.6251,
|
||||||
|
"eval_samples_per_second": 45.535,
|
||||||
|
"eval_steps_per_second": 1.846,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8452938675880433,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"grad_norm": 0.789717972278595,
|
||||||
|
"learning_rate": 0.00023816778784387094,
|
||||||
|
"loss": 0.8828925132751465,
|
||||||
|
"mean_token_accuracy": 0.8208261251449585,
|
||||||
|
"num_tokens": 530794.0,
|
||||||
|
"step": 45
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5805854141712189,
|
||||||
|
"epoch": 1.1111111111111112,
|
||||||
|
"grad_norm": 0.4138790965080261,
|
||||||
|
"learning_rate": 0.00022289751198639087,
|
||||||
|
"loss": 0.5360542297363281,
|
||||||
|
"mean_token_accuracy": 0.8808212757110596,
|
||||||
|
"num_tokens": 589773.0,
|
||||||
|
"step": 50
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8660945177078248,
|
||||||
|
"epoch": 1.2222222222222223,
|
||||||
|
"grad_norm": 0.4917871356010437,
|
||||||
|
"learning_rate": 0.00020656422625324805,
|
||||||
|
"loss": 0.8303131103515625,
|
||||||
|
"mean_token_accuracy": 0.8197528839111328,
|
||||||
|
"num_tokens": 644861.0,
|
||||||
|
"step": 55
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5908387929201127,
|
||||||
|
"epoch": 1.3333333333333333,
|
||||||
|
"grad_norm": 0.5149257779121399,
|
||||||
|
"learning_rate": 0.00018940610672978802,
|
||||||
|
"loss": 0.5536936283111572,
|
||||||
|
"mean_token_accuracy": 0.8759726524353028,
|
||||||
|
"num_tokens": 707064.0,
|
||||||
|
"step": 60
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.3333333333333333,
|
||||||
|
"eval_entropy": 0.7562275131543478,
|
||||||
|
"eval_loss": 0.9295161366462708,
|
||||||
|
"eval_mean_token_accuracy": 0.8002562522888184,
|
||||||
|
"eval_num_tokens": 707064.0,
|
||||||
|
"eval_runtime": 1.6273,
|
||||||
|
"eval_samples_per_second": 45.475,
|
||||||
|
"eval_steps_per_second": 1.844,
|
||||||
|
"step": 60
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7860998153686524,
|
||||||
|
"epoch": 1.4444444444444444,
|
||||||
|
"grad_norm": 0.6012740731239319,
|
||||||
|
"learning_rate": 0.00017167335743538317,
|
||||||
|
"loss": 0.7982378482818604,
|
||||||
|
"mean_token_accuracy": 0.8309973835945129,
|
||||||
|
"num_tokens": 763959.0,
|
||||||
|
"step": 65
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6057170808315278,
|
||||||
|
"epoch": 1.5555555555555556,
|
||||||
|
"grad_norm": 0.7193827629089355,
|
||||||
|
"learning_rate": 0.00015362456178541982,
|
||||||
|
"loss": 0.6322322368621827,
|
||||||
|
"mean_token_accuracy": 0.8604895114898682,
|
||||||
|
"num_tokens": 826139.0,
|
||||||
|
"step": 70
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8781150817871094,
|
||||||
|
"epoch": 1.6666666666666665,
|
||||||
|
"grad_norm": 0.5556202530860901,
|
||||||
|
"learning_rate": 0.00013552291186282273,
|
||||||
|
"loss": 0.8648699760437012,
|
||||||
|
"mean_token_accuracy": 0.8099217295646668,
|
||||||
|
"num_tokens": 883964.0,
|
||||||
|
"step": 75
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7820337891578675,
|
||||||
|
"epoch": 1.7777777777777777,
|
||||||
|
"grad_norm": 0.626190721988678,
|
||||||
|
"learning_rate": 0.00011763237048482908,
|
||||||
|
"loss": 0.7500524520874023,
|
||||||
|
"mean_token_accuracy": 0.8313877940177917,
|
||||||
|
"num_tokens": 942825.0,
|
||||||
|
"step": 80
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.7777777777777777,
|
||||||
|
"eval_entropy": 0.7708820303281149,
|
||||||
|
"eval_loss": 0.9428374767303467,
|
||||||
|
"eval_mean_token_accuracy": 0.7990738352139791,
|
||||||
|
"eval_num_tokens": 942825.0,
|
||||||
|
"eval_runtime": 1.6267,
|
||||||
|
"eval_samples_per_second": 45.49,
|
||||||
|
"eval_steps_per_second": 1.844,
|
||||||
|
"step": 80
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6434245944023133,
|
||||||
|
"epoch": 1.8888888888888888,
|
||||||
|
"grad_norm": 0.7037157416343689,
|
||||||
|
"learning_rate": 0.00010021382203072075,
|
||||||
|
"loss": 0.6138159275054932,
|
||||||
|
"mean_token_accuracy": 0.8665528893470764,
|
||||||
|
"num_tokens": 1005474.0,
|
||||||
|
"step": 85
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.44103629887104034,
|
||||||
|
"epoch": 2.0,
|
||||||
|
"grad_norm": 0.5344287753105164,
|
||||||
|
"learning_rate": 8.352126816011381e-05,
|
||||||
|
"loss": 0.441878604888916,
|
||||||
|
"mean_token_accuracy": 0.9036641597747803,
|
||||||
|
"num_tokens": 1061588.0,
|
||||||
|
"step": 90
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7921009421348572,
|
||||||
|
"epoch": 2.111111111111111,
|
||||||
|
"grad_norm": 0.5275471210479736,
|
||||||
|
"learning_rate": 6.779812389679952e-05,
|
||||||
|
"loss": 0.7220367431640625,
|
||||||
|
"mean_token_accuracy": 0.8374058127403259,
|
||||||
|
"num_tokens": 1119675.0,
|
||||||
|
"step": 95
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6127014696598053,
|
||||||
|
"epoch": 2.2222222222222223,
|
||||||
|
"grad_norm": 0.6188642382621765,
|
||||||
|
"learning_rate": 5.327366808957444e-05,
|
||||||
|
"loss": 0.545912790298462,
|
||||||
|
"mean_token_accuracy": 0.8728284955024719,
|
||||||
|
"num_tokens": 1178607.0,
|
||||||
|
"step": 100
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.2222222222222223,
|
||||||
|
"eval_entropy": 0.6974611083666483,
|
||||||
|
"eval_loss": 0.9652960300445557,
|
||||||
|
"eval_mean_token_accuracy": 0.799143652121226,
|
||||||
|
"eval_num_tokens": 1178607.0,
|
||||||
|
"eval_runtime": 1.6275,
|
||||||
|
"eval_samples_per_second": 45.469,
|
||||||
|
"eval_steps_per_second": 1.843,
|
||||||
|
"step": 100
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5584009617567063,
|
||||||
|
"epoch": 2.3333333333333335,
|
||||||
|
"grad_norm": 0.2800925076007843,
|
||||||
|
"learning_rate": 4.015970001033401e-05,
|
||||||
|
"loss": 0.5415268421173096,
|
||||||
|
"mean_token_accuracy": 0.8769876599311829,
|
||||||
|
"num_tokens": 1239996.0,
|
||||||
|
"step": 105
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.40896390080451966,
|
||||||
|
"epoch": 2.4444444444444446,
|
||||||
|
"grad_norm": 0.38913053274154663,
|
||||||
|
"learning_rate": 2.8647450843757897e-05,
|
||||||
|
"loss": 0.4072125434875488,
|
||||||
|
"mean_token_accuracy": 0.9056331634521484,
|
||||||
|
"num_tokens": 1303203.0,
|
||||||
|
"step": 110
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5653573453426362,
|
||||||
|
"epoch": 2.5555555555555554,
|
||||||
|
"grad_norm": 0.7067868709564209,
|
||||||
|
"learning_rate": 1.8904795106022313e-05,
|
||||||
|
"loss": 0.5507754325866699,
|
||||||
|
"mean_token_accuracy": 0.8763858556747437,
|
||||||
|
"num_tokens": 1361238.0,
|
||||||
|
"step": 115
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5153138220310212,
|
||||||
|
"epoch": 2.6666666666666665,
|
||||||
|
"grad_norm": 0.7621523141860962,
|
||||||
|
"learning_rate": 1.1073802656333546e-05,
|
||||||
|
"loss": 0.507559871673584,
|
||||||
|
"mean_token_accuracy": 0.8800268650054932,
|
||||||
|
"num_tokens": 1423474.0,
|
||||||
|
"step": 120
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 2.6666666666666665,
|
||||||
|
"eval_entropy": 0.6664723753929138,
|
||||||
|
"eval_loss": 0.9865095615386963,
|
||||||
|
"eval_mean_token_accuracy": 0.7970906893412272,
|
||||||
|
"eval_num_tokens": 1423474.0,
|
||||||
|
"eval_runtime": 1.6278,
|
||||||
|
"eval_samples_per_second": 45.46,
|
||||||
|
"eval_steps_per_second": 1.843,
|
||||||
|
"step": 120
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.634662801027298,
|
||||||
|
"epoch": 2.7777777777777777,
|
||||||
|
"grad_norm": 0.5899237394332886,
|
||||||
|
"learning_rate": 5.268666998467297e-06,
|
||||||
|
"loss": 0.5941447734832763,
|
||||||
|
"mean_token_accuracy": 0.8620630621910095,
|
||||||
|
"num_tokens": 1479493.0,
|
||||||
|
"step": 125
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5771006941795349,
|
||||||
|
"epoch": 2.888888888888889,
|
||||||
|
"grad_norm": 0.6653247475624084,
|
||||||
|
"learning_rate": 1.5740400823423316e-06,
|
||||||
|
"loss": 0.5706047534942627,
|
||||||
|
"mean_token_accuracy": 0.874986720085144,
|
||||||
|
"num_tokens": 1539267.0,
|
||||||
|
"step": 130
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.603338897228241,
|
||||||
|
"epoch": 3.0,
|
||||||
|
"grad_norm": 0.9464746713638306,
|
||||||
|
"learning_rate": 4.379788797105877e-08,
|
||||||
|
"loss": 0.5793191909790039,
|
||||||
|
"mean_token_accuracy": 0.8678125262260437,
|
||||||
|
"num_tokens": 1592382.0,
|
||||||
|
"step": 135
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 3.0,
|
||||||
|
"eval_entropy": 0.6672613223393759,
|
||||||
|
"eval_loss": 0.9868772625923157,
|
||||||
|
"eval_mean_token_accuracy": 0.7981111407279968,
|
||||||
|
"eval_num_tokens": 1592382.0,
|
||||||
|
"eval_runtime": 1.6294,
|
||||||
|
"eval_samples_per_second": 45.415,
|
||||||
|
"eval_steps_per_second": 1.841,
|
||||||
|
"step": 135
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"logging_steps": 5,
|
||||||
|
"max_steps": 135,
|
||||||
|
"num_input_tokens_seen": 0,
|
||||||
|
"num_train_epochs": 3,
|
||||||
|
"save_steps": 500,
|
||||||
|
"stateful_callbacks": {
|
||||||
|
"TrainerControl": {
|
||||||
|
"args": {
|
||||||
|
"should_epoch_stop": false,
|
||||||
|
"should_evaluate": false,
|
||||||
|
"should_log": false,
|
||||||
|
"should_save": true,
|
||||||
|
"should_training_stop": true
|
||||||
|
},
|
||||||
|
"attributes": {}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"total_flos": 5851882291200000.0,
|
||||||
|
"train_batch_size": 32,
|
||||||
|
"trial_name": null,
|
||||||
|
"trial_params": null
|
||||||
|
}
|
||||||
3
adapters/sft-lora/checkpoint-135/training_args.bin
Normal file
3
adapters/sft-lora/checkpoint-135/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:634fa3e1efca4fb9cf47831d4f352701147dd578361c76a8defa5309a5ed79d4
|
||||||
|
size 5713
|
||||||
209
adapters/sft-lora/checkpoint-45/README.md
Normal file
209
adapters/sft-lora/checkpoint-45/README.md
Normal file
@@ -0,0 +1,209 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for Model ID
|
||||||
|
|
||||||
|
<!-- Provide a quick summary of what the model is/does. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Details
|
||||||
|
|
||||||
|
### Model Description
|
||||||
|
|
||||||
|
<!-- Provide a longer summary of what this model is. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
- **Developed by:** [More Information Needed]
|
||||||
|
- **Funded by [optional]:** [More Information Needed]
|
||||||
|
- **Shared by [optional]:** [More Information Needed]
|
||||||
|
- **Model type:** [More Information Needed]
|
||||||
|
- **Language(s) (NLP):** [More Information Needed]
|
||||||
|
- **License:** [More Information Needed]
|
||||||
|
- **Finetuned from model [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
### Model Sources [optional]
|
||||||
|
|
||||||
|
<!-- Provide the basic links for the model. -->
|
||||||
|
|
||||||
|
- **Repository:** [More Information Needed]
|
||||||
|
- **Paper [optional]:** [More Information Needed]
|
||||||
|
- **Demo [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
## Uses
|
||||||
|
|
||||||
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
||||||
|
|
||||||
|
### Direct Use
|
||||||
|
|
||||||
|
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Downstream Use [optional]
|
||||||
|
|
||||||
|
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Out-of-Scope Use
|
||||||
|
|
||||||
|
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Bias, Risks, and Limitations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Recommendations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
||||||
|
|
||||||
|
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
||||||
|
|
||||||
|
## How to Get Started with the Model
|
||||||
|
|
||||||
|
Use the code below to get started with the model.
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Training Details
|
||||||
|
|
||||||
|
### Training Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Training Procedure
|
||||||
|
|
||||||
|
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
||||||
|
|
||||||
|
#### Preprocessing [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
|
||||||
|
#### Training Hyperparameters
|
||||||
|
|
||||||
|
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
||||||
|
|
||||||
|
#### Speeds, Sizes, Times [optional]
|
||||||
|
|
||||||
|
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Evaluation
|
||||||
|
|
||||||
|
<!-- This section describes the evaluation protocols and provides the results. -->
|
||||||
|
|
||||||
|
### Testing Data, Factors & Metrics
|
||||||
|
|
||||||
|
#### Testing Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card if possible. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Factors
|
||||||
|
|
||||||
|
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Metrics
|
||||||
|
|
||||||
|
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Results
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Summary
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Examination [optional]
|
||||||
|
|
||||||
|
<!-- Relevant interpretability work for the model goes here -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Environmental Impact
|
||||||
|
|
||||||
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
||||||
|
|
||||||
|
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
||||||
|
|
||||||
|
- **Hardware Type:** [More Information Needed]
|
||||||
|
- **Hours used:** [More Information Needed]
|
||||||
|
- **Cloud Provider:** [More Information Needed]
|
||||||
|
- **Compute Region:** [More Information Needed]
|
||||||
|
- **Carbon Emitted:** [More Information Needed]
|
||||||
|
|
||||||
|
## Technical Specifications [optional]
|
||||||
|
|
||||||
|
### Model Architecture and Objective
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Compute Infrastructure
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Hardware
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Software
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Citation [optional]
|
||||||
|
|
||||||
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
||||||
|
|
||||||
|
**BibTeX:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
**APA:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Glossary [optional]
|
||||||
|
|
||||||
|
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## More Information [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Authors [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Contact
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
48
adapters/sft-lora/checkpoint-45/adapter_config.json
Normal file
48
adapters/sft-lora/checkpoint-45/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"gate_proj",
|
||||||
|
"down_proj",
|
||||||
|
"o_proj",
|
||||||
|
"v_proj",
|
||||||
|
"k_proj",
|
||||||
|
"up_proj",
|
||||||
|
"q_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:92c05a7c7a5df4f7933be198d7f325b08b4eb5734df7f73d5ef68beb112b78ad
|
||||||
|
size 40422168
|
||||||
89
adapters/sft-lora/checkpoint-45/chat_template.jinja
Normal file
89
adapters/sft-lora/checkpoint-45/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
3
adapters/sft-lora/checkpoint-45/optimizer.pt
Normal file
3
adapters/sft-lora/checkpoint-45/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:9eccb6b9f7f8006c3819e8ec2be7a4bac4e85a31f3e9ce1fba861f2afa55f33c
|
||||||
|
size 81075515
|
||||||
3
adapters/sft-lora/checkpoint-45/rng_state.pth
Normal file
3
adapters/sft-lora/checkpoint-45/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:ddd57472f391fa9a4127029522f710d7e0ed0d23bac401252dfd41baf889ec70
|
||||||
|
size 14645
|
||||||
3
adapters/sft-lora/checkpoint-45/scheduler.pt
Normal file
3
adapters/sft-lora/checkpoint-45/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:b42498054c7029f5eee8f8a1a55d853c5e20f855fa41160c0f21019d45eae999
|
||||||
|
size 1465
|
||||||
3
adapters/sft-lora/checkpoint-45/tokenizer.json
Normal file
3
adapters/sft-lora/checkpoint-45/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sft-lora/checkpoint-45/tokenizer_config.json
Normal file
30
adapters/sft-lora/checkpoint-45/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
146
adapters/sft-lora/checkpoint-45/trainer_state.json
Normal file
146
adapters/sft-lora/checkpoint-45/trainer_state.json
Normal file
@@ -0,0 +1,146 @@
|
|||||||
|
{
|
||||||
|
"best_global_step": null,
|
||||||
|
"best_metric": null,
|
||||||
|
"best_model_checkpoint": null,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"eval_steps": 20,
|
||||||
|
"global_step": 45,
|
||||||
|
"is_hyper_param_search": false,
|
||||||
|
"is_local_process_zero": true,
|
||||||
|
"is_world_process_zero": true,
|
||||||
|
"log_history": [
|
||||||
|
{
|
||||||
|
"entropy": 0.46982616782188413,
|
||||||
|
"epoch": 0.1111111111111111,
|
||||||
|
"grad_norm": 1.9415351152420044,
|
||||||
|
"learning_rate": 0.00023999999999999998,
|
||||||
|
"loss": 1.166135025024414,
|
||||||
|
"mean_token_accuracy": 0.8076520800590515,
|
||||||
|
"num_tokens": 59002.0,
|
||||||
|
"step": 5
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.681049507856369,
|
||||||
|
"epoch": 0.2222222222222222,
|
||||||
|
"grad_norm": 0.8071348071098328,
|
||||||
|
"learning_rate": 0.00029929974520764727,
|
||||||
|
"loss": 0.7030046463012696,
|
||||||
|
"mean_token_accuracy": 0.849864411354065,
|
||||||
|
"num_tokens": 120432.0,
|
||||||
|
"step": 10
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.9322404026985168,
|
||||||
|
"epoch": 0.3333333333333333,
|
||||||
|
"grad_norm": 0.6664227247238159,
|
||||||
|
"learning_rate": 0.0002964661618243908,
|
||||||
|
"loss": 0.878734016418457,
|
||||||
|
"mean_token_accuracy": 0.8135401248931885,
|
||||||
|
"num_tokens": 179678.0,
|
||||||
|
"step": 15
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8074370503425599,
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"grad_norm": 0.6451402306556702,
|
||||||
|
"learning_rate": 0.0002914967719886614,
|
||||||
|
"loss": 0.8701633453369141,
|
||||||
|
"mean_token_accuracy": 0.8204897165298461,
|
||||||
|
"num_tokens": 240600.0,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"eval_entropy": 0.8143169482549032,
|
||||||
|
"eval_loss": 0.924848735332489,
|
||||||
|
"eval_mean_token_accuracy": 0.803371528784434,
|
||||||
|
"eval_num_tokens": 240600.0,
|
||||||
|
"eval_runtime": 1.6201,
|
||||||
|
"eval_samples_per_second": 45.676,
|
||||||
|
"eval_steps_per_second": 1.852,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7912694752216339,
|
||||||
|
"epoch": 0.5555555555555556,
|
||||||
|
"grad_norm": 0.4144524037837982,
|
||||||
|
"learning_rate": 0.0002844640405943555,
|
||||||
|
"loss": 0.8696941375732422,
|
||||||
|
"mean_token_accuracy": 0.8263819813728333,
|
||||||
|
"num_tokens": 301638.0,
|
||||||
|
"step": 25
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.9237359523773193,
|
||||||
|
"epoch": 0.6666666666666666,
|
||||||
|
"grad_norm": 0.5862515568733215,
|
||||||
|
"learning_rate": 0.00027547052070153396,
|
||||||
|
"loss": 0.9361928939819336,
|
||||||
|
"mean_token_accuracy": 0.8028241634368897,
|
||||||
|
"num_tokens": 359243.0,
|
||||||
|
"step": 30
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6837044417858124,
|
||||||
|
"epoch": 0.7777777777777778,
|
||||||
|
"grad_norm": 0.4594927728176117,
|
||||||
|
"learning_rate": 0.0002646473580818772,
|
||||||
|
"loss": 0.6992801666259766,
|
||||||
|
"mean_token_accuracy": 0.8513208270072937,
|
||||||
|
"num_tokens": 417098.0,
|
||||||
|
"step": 35
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8795680403709412,
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"grad_norm": 0.5710228681564331,
|
||||||
|
"learning_rate": 0.00025215237881801955,
|
||||||
|
"loss": 0.9085055351257324,
|
||||||
|
"mean_token_accuracy": 0.8098101854324341,
|
||||||
|
"num_tokens": 480008.0,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"eval_entropy": 0.8618650635083517,
|
||||||
|
"eval_loss": 0.9124361872673035,
|
||||||
|
"eval_mean_token_accuracy": 0.8028692404429117,
|
||||||
|
"eval_num_tokens": 480008.0,
|
||||||
|
"eval_runtime": 1.6251,
|
||||||
|
"eval_samples_per_second": 45.535,
|
||||||
|
"eval_steps_per_second": 1.846,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8452938675880433,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"grad_norm": 0.789717972278595,
|
||||||
|
"learning_rate": 0.00023816778784387094,
|
||||||
|
"loss": 0.8828925132751465,
|
||||||
|
"mean_token_accuracy": 0.8208261251449585,
|
||||||
|
"num_tokens": 530794.0,
|
||||||
|
"step": 45
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"logging_steps": 5,
|
||||||
|
"max_steps": 135,
|
||||||
|
"num_input_tokens_seen": 0,
|
||||||
|
"num_train_epochs": 3,
|
||||||
|
"save_steps": 500,
|
||||||
|
"stateful_callbacks": {
|
||||||
|
"TrainerControl": {
|
||||||
|
"args": {
|
||||||
|
"should_epoch_stop": false,
|
||||||
|
"should_evaluate": false,
|
||||||
|
"should_log": false,
|
||||||
|
"should_save": true,
|
||||||
|
"should_training_stop": false
|
||||||
|
},
|
||||||
|
"attributes": {}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"total_flos": 1952299008000000.0,
|
||||||
|
"train_batch_size": 32,
|
||||||
|
"trial_name": null,
|
||||||
|
"trial_params": null
|
||||||
|
}
|
||||||
3
adapters/sft-lora/checkpoint-45/training_args.bin
Normal file
3
adapters/sft-lora/checkpoint-45/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:634fa3e1efca4fb9cf47831d4f352701147dd578361c76a8defa5309a5ed79d4
|
||||||
|
size 5713
|
||||||
209
adapters/sft-lora/checkpoint-90/README.md
Normal file
209
adapters/sft-lora/checkpoint-90/README.md
Normal file
@@ -0,0 +1,209 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for Model ID
|
||||||
|
|
||||||
|
<!-- Provide a quick summary of what the model is/does. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Details
|
||||||
|
|
||||||
|
### Model Description
|
||||||
|
|
||||||
|
<!-- Provide a longer summary of what this model is. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
- **Developed by:** [More Information Needed]
|
||||||
|
- **Funded by [optional]:** [More Information Needed]
|
||||||
|
- **Shared by [optional]:** [More Information Needed]
|
||||||
|
- **Model type:** [More Information Needed]
|
||||||
|
- **Language(s) (NLP):** [More Information Needed]
|
||||||
|
- **License:** [More Information Needed]
|
||||||
|
- **Finetuned from model [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
### Model Sources [optional]
|
||||||
|
|
||||||
|
<!-- Provide the basic links for the model. -->
|
||||||
|
|
||||||
|
- **Repository:** [More Information Needed]
|
||||||
|
- **Paper [optional]:** [More Information Needed]
|
||||||
|
- **Demo [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
## Uses
|
||||||
|
|
||||||
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
||||||
|
|
||||||
|
### Direct Use
|
||||||
|
|
||||||
|
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Downstream Use [optional]
|
||||||
|
|
||||||
|
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Out-of-Scope Use
|
||||||
|
|
||||||
|
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Bias, Risks, and Limitations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Recommendations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
||||||
|
|
||||||
|
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
||||||
|
|
||||||
|
## How to Get Started with the Model
|
||||||
|
|
||||||
|
Use the code below to get started with the model.
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Training Details
|
||||||
|
|
||||||
|
### Training Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Training Procedure
|
||||||
|
|
||||||
|
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
||||||
|
|
||||||
|
#### Preprocessing [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
|
||||||
|
#### Training Hyperparameters
|
||||||
|
|
||||||
|
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
||||||
|
|
||||||
|
#### Speeds, Sizes, Times [optional]
|
||||||
|
|
||||||
|
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Evaluation
|
||||||
|
|
||||||
|
<!-- This section describes the evaluation protocols and provides the results. -->
|
||||||
|
|
||||||
|
### Testing Data, Factors & Metrics
|
||||||
|
|
||||||
|
#### Testing Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card if possible. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Factors
|
||||||
|
|
||||||
|
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Metrics
|
||||||
|
|
||||||
|
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Results
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Summary
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Examination [optional]
|
||||||
|
|
||||||
|
<!-- Relevant interpretability work for the model goes here -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Environmental Impact
|
||||||
|
|
||||||
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
||||||
|
|
||||||
|
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
||||||
|
|
||||||
|
- **Hardware Type:** [More Information Needed]
|
||||||
|
- **Hours used:** [More Information Needed]
|
||||||
|
- **Cloud Provider:** [More Information Needed]
|
||||||
|
- **Compute Region:** [More Information Needed]
|
||||||
|
- **Carbon Emitted:** [More Information Needed]
|
||||||
|
|
||||||
|
## Technical Specifications [optional]
|
||||||
|
|
||||||
|
### Model Architecture and Objective
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Compute Infrastructure
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Hardware
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Software
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Citation [optional]
|
||||||
|
|
||||||
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
||||||
|
|
||||||
|
**BibTeX:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
**APA:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Glossary [optional]
|
||||||
|
|
||||||
|
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## More Information [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Authors [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Contact
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
48
adapters/sft-lora/checkpoint-90/adapter_config.json
Normal file
48
adapters/sft-lora/checkpoint-90/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"gate_proj",
|
||||||
|
"down_proj",
|
||||||
|
"o_proj",
|
||||||
|
"v_proj",
|
||||||
|
"k_proj",
|
||||||
|
"up_proj",
|
||||||
|
"q_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:f9b7aafe23996e555ee3c633855ded62f0d3a41827f7a2947126239a300cca06
|
||||||
|
size 40422168
|
||||||
89
adapters/sft-lora/checkpoint-90/chat_template.jinja
Normal file
89
adapters/sft-lora/checkpoint-90/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
3
adapters/sft-lora/checkpoint-90/optimizer.pt
Normal file
3
adapters/sft-lora/checkpoint-90/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:c02077aa2d62c8b071fbdc0157e2ab15d2e8f04fb2ff5bf8bef27f7399f5f873
|
||||||
|
size 81075515
|
||||||
3
adapters/sft-lora/checkpoint-90/rng_state.pth
Normal file
3
adapters/sft-lora/checkpoint-90/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:a29e79e95308c75e7702cb8bb76ee726dec3cc4986dacfcb847c4ad283108edd
|
||||||
|
size 14645
|
||||||
3
adapters/sft-lora/checkpoint-90/scheduler.pt
Normal file
3
adapters/sft-lora/checkpoint-90/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:d8f6b5fec4d818c43b5627862d205d6de3c227cbb6841c5d3cebd3072b69491f
|
||||||
|
size 1465
|
||||||
3
adapters/sft-lora/checkpoint-90/tokenizer.json
Normal file
3
adapters/sft-lora/checkpoint-90/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sft-lora/checkpoint-90/tokenizer_config.json
Normal file
30
adapters/sft-lora/checkpoint-90/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
258
adapters/sft-lora/checkpoint-90/trainer_state.json
Normal file
258
adapters/sft-lora/checkpoint-90/trainer_state.json
Normal file
@@ -0,0 +1,258 @@
|
|||||||
|
{
|
||||||
|
"best_global_step": null,
|
||||||
|
"best_metric": null,
|
||||||
|
"best_model_checkpoint": null,
|
||||||
|
"epoch": 2.0,
|
||||||
|
"eval_steps": 20,
|
||||||
|
"global_step": 90,
|
||||||
|
"is_hyper_param_search": false,
|
||||||
|
"is_local_process_zero": true,
|
||||||
|
"is_world_process_zero": true,
|
||||||
|
"log_history": [
|
||||||
|
{
|
||||||
|
"entropy": 0.46982616782188413,
|
||||||
|
"epoch": 0.1111111111111111,
|
||||||
|
"grad_norm": 1.9415351152420044,
|
||||||
|
"learning_rate": 0.00023999999999999998,
|
||||||
|
"loss": 1.166135025024414,
|
||||||
|
"mean_token_accuracy": 0.8076520800590515,
|
||||||
|
"num_tokens": 59002.0,
|
||||||
|
"step": 5
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.681049507856369,
|
||||||
|
"epoch": 0.2222222222222222,
|
||||||
|
"grad_norm": 0.8071348071098328,
|
||||||
|
"learning_rate": 0.00029929974520764727,
|
||||||
|
"loss": 0.7030046463012696,
|
||||||
|
"mean_token_accuracy": 0.849864411354065,
|
||||||
|
"num_tokens": 120432.0,
|
||||||
|
"step": 10
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.9322404026985168,
|
||||||
|
"epoch": 0.3333333333333333,
|
||||||
|
"grad_norm": 0.6664227247238159,
|
||||||
|
"learning_rate": 0.0002964661618243908,
|
||||||
|
"loss": 0.878734016418457,
|
||||||
|
"mean_token_accuracy": 0.8135401248931885,
|
||||||
|
"num_tokens": 179678.0,
|
||||||
|
"step": 15
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8074370503425599,
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"grad_norm": 0.6451402306556702,
|
||||||
|
"learning_rate": 0.0002914967719886614,
|
||||||
|
"loss": 0.8701633453369141,
|
||||||
|
"mean_token_accuracy": 0.8204897165298461,
|
||||||
|
"num_tokens": 240600.0,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4444444444444444,
|
||||||
|
"eval_entropy": 0.8143169482549032,
|
||||||
|
"eval_loss": 0.924848735332489,
|
||||||
|
"eval_mean_token_accuracy": 0.803371528784434,
|
||||||
|
"eval_num_tokens": 240600.0,
|
||||||
|
"eval_runtime": 1.6201,
|
||||||
|
"eval_samples_per_second": 45.676,
|
||||||
|
"eval_steps_per_second": 1.852,
|
||||||
|
"step": 20
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7912694752216339,
|
||||||
|
"epoch": 0.5555555555555556,
|
||||||
|
"grad_norm": 0.4144524037837982,
|
||||||
|
"learning_rate": 0.0002844640405943555,
|
||||||
|
"loss": 0.8696941375732422,
|
||||||
|
"mean_token_accuracy": 0.8263819813728333,
|
||||||
|
"num_tokens": 301638.0,
|
||||||
|
"step": 25
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.9237359523773193,
|
||||||
|
"epoch": 0.6666666666666666,
|
||||||
|
"grad_norm": 0.5862515568733215,
|
||||||
|
"learning_rate": 0.00027547052070153396,
|
||||||
|
"loss": 0.9361928939819336,
|
||||||
|
"mean_token_accuracy": 0.8028241634368897,
|
||||||
|
"num_tokens": 359243.0,
|
||||||
|
"step": 30
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6837044417858124,
|
||||||
|
"epoch": 0.7777777777777778,
|
||||||
|
"grad_norm": 0.4594927728176117,
|
||||||
|
"learning_rate": 0.0002646473580818772,
|
||||||
|
"loss": 0.6992801666259766,
|
||||||
|
"mean_token_accuracy": 0.8513208270072937,
|
||||||
|
"num_tokens": 417098.0,
|
||||||
|
"step": 35
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8795680403709412,
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"grad_norm": 0.5710228681564331,
|
||||||
|
"learning_rate": 0.00025215237881801955,
|
||||||
|
"loss": 0.9085055351257324,
|
||||||
|
"mean_token_accuracy": 0.8098101854324341,
|
||||||
|
"num_tokens": 480008.0,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.8888888888888888,
|
||||||
|
"eval_entropy": 0.8618650635083517,
|
||||||
|
"eval_loss": 0.9124361872673035,
|
||||||
|
"eval_mean_token_accuracy": 0.8028692404429117,
|
||||||
|
"eval_num_tokens": 480008.0,
|
||||||
|
"eval_runtime": 1.6251,
|
||||||
|
"eval_samples_per_second": 45.535,
|
||||||
|
"eval_steps_per_second": 1.846,
|
||||||
|
"step": 40
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8452938675880433,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"grad_norm": 0.789717972278595,
|
||||||
|
"learning_rate": 0.00023816778784387094,
|
||||||
|
"loss": 0.8828925132751465,
|
||||||
|
"mean_token_accuracy": 0.8208261251449585,
|
||||||
|
"num_tokens": 530794.0,
|
||||||
|
"step": 45
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5805854141712189,
|
||||||
|
"epoch": 1.1111111111111112,
|
||||||
|
"grad_norm": 0.4138790965080261,
|
||||||
|
"learning_rate": 0.00022289751198639087,
|
||||||
|
"loss": 0.5360542297363281,
|
||||||
|
"mean_token_accuracy": 0.8808212757110596,
|
||||||
|
"num_tokens": 589773.0,
|
||||||
|
"step": 50
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8660945177078248,
|
||||||
|
"epoch": 1.2222222222222223,
|
||||||
|
"grad_norm": 0.4917871356010437,
|
||||||
|
"learning_rate": 0.00020656422625324805,
|
||||||
|
"loss": 0.8303131103515625,
|
||||||
|
"mean_token_accuracy": 0.8197528839111328,
|
||||||
|
"num_tokens": 644861.0,
|
||||||
|
"step": 55
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5908387929201127,
|
||||||
|
"epoch": 1.3333333333333333,
|
||||||
|
"grad_norm": 0.5149257779121399,
|
||||||
|
"learning_rate": 0.00018940610672978802,
|
||||||
|
"loss": 0.5536936283111572,
|
||||||
|
"mean_token_accuracy": 0.8759726524353028,
|
||||||
|
"num_tokens": 707064.0,
|
||||||
|
"step": 60
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.3333333333333333,
|
||||||
|
"eval_entropy": 0.7562275131543478,
|
||||||
|
"eval_loss": 0.9295161366462708,
|
||||||
|
"eval_mean_token_accuracy": 0.8002562522888184,
|
||||||
|
"eval_num_tokens": 707064.0,
|
||||||
|
"eval_runtime": 1.6273,
|
||||||
|
"eval_samples_per_second": 45.475,
|
||||||
|
"eval_steps_per_second": 1.844,
|
||||||
|
"step": 60
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7860998153686524,
|
||||||
|
"epoch": 1.4444444444444444,
|
||||||
|
"grad_norm": 0.6012740731239319,
|
||||||
|
"learning_rate": 0.00017167335743538317,
|
||||||
|
"loss": 0.7982378482818604,
|
||||||
|
"mean_token_accuracy": 0.8309973835945129,
|
||||||
|
"num_tokens": 763959.0,
|
||||||
|
"step": 65
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6057170808315278,
|
||||||
|
"epoch": 1.5555555555555556,
|
||||||
|
"grad_norm": 0.7193827629089355,
|
||||||
|
"learning_rate": 0.00015362456178541982,
|
||||||
|
"loss": 0.6322322368621827,
|
||||||
|
"mean_token_accuracy": 0.8604895114898682,
|
||||||
|
"num_tokens": 826139.0,
|
||||||
|
"step": 70
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.8781150817871094,
|
||||||
|
"epoch": 1.6666666666666665,
|
||||||
|
"grad_norm": 0.5556202530860901,
|
||||||
|
"learning_rate": 0.00013552291186282273,
|
||||||
|
"loss": 0.8648699760437012,
|
||||||
|
"mean_token_accuracy": 0.8099217295646668,
|
||||||
|
"num_tokens": 883964.0,
|
||||||
|
"step": 75
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7820337891578675,
|
||||||
|
"epoch": 1.7777777777777777,
|
||||||
|
"grad_norm": 0.626190721988678,
|
||||||
|
"learning_rate": 0.00011763237048482908,
|
||||||
|
"loss": 0.7500524520874023,
|
||||||
|
"mean_token_accuracy": 0.8313877940177917,
|
||||||
|
"num_tokens": 942825.0,
|
||||||
|
"step": 80
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.7777777777777777,
|
||||||
|
"eval_entropy": 0.7708820303281149,
|
||||||
|
"eval_loss": 0.9428374767303467,
|
||||||
|
"eval_mean_token_accuracy": 0.7990738352139791,
|
||||||
|
"eval_num_tokens": 942825.0,
|
||||||
|
"eval_runtime": 1.6267,
|
||||||
|
"eval_samples_per_second": 45.49,
|
||||||
|
"eval_steps_per_second": 1.844,
|
||||||
|
"step": 80
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.6434245944023133,
|
||||||
|
"epoch": 1.8888888888888888,
|
||||||
|
"grad_norm": 0.7037157416343689,
|
||||||
|
"learning_rate": 0.00010021382203072075,
|
||||||
|
"loss": 0.6138159275054932,
|
||||||
|
"mean_token_accuracy": 0.8665528893470764,
|
||||||
|
"num_tokens": 1005474.0,
|
||||||
|
"step": 85
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.44103629887104034,
|
||||||
|
"epoch": 2.0,
|
||||||
|
"grad_norm": 0.5344287753105164,
|
||||||
|
"learning_rate": 8.352126816011381e-05,
|
||||||
|
"loss": 0.441878604888916,
|
||||||
|
"mean_token_accuracy": 0.9036641597747803,
|
||||||
|
"num_tokens": 1061588.0,
|
||||||
|
"step": 90
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"logging_steps": 5,
|
||||||
|
"max_steps": 135,
|
||||||
|
"num_input_tokens_seen": 0,
|
||||||
|
"num_train_epochs": 3,
|
||||||
|
"save_steps": 500,
|
||||||
|
"stateful_callbacks": {
|
||||||
|
"TrainerControl": {
|
||||||
|
"args": {
|
||||||
|
"should_epoch_stop": false,
|
||||||
|
"should_evaluate": false,
|
||||||
|
"should_log": false,
|
||||||
|
"should_save": true,
|
||||||
|
"should_training_stop": false
|
||||||
|
},
|
||||||
|
"attributes": {}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"total_flos": 3901053911040000.0,
|
||||||
|
"train_batch_size": 32,
|
||||||
|
"trial_name": null,
|
||||||
|
"trial_params": null
|
||||||
|
}
|
||||||
3
adapters/sft-lora/checkpoint-90/training_args.bin
Normal file
3
adapters/sft-lora/checkpoint-90/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:634fa3e1efca4fb9cf47831d4f352701147dd578361c76a8defa5309a5ed79d4
|
||||||
|
size 5713
|
||||||
3
adapters/sft-lora/tokenizer.json
Normal file
3
adapters/sft-lora/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sft-lora/tokenizer_config.json
Normal file
30
adapters/sft-lora/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
3
adapters/sft-lora/training_args.bin
Normal file
3
adapters/sft-lora/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:634fa3e1efca4fb9cf47831d4f352701147dd578361c76a8defa5309a5ed79d4
|
||||||
|
size 5713
|
||||||
62
adapters/sweep-1e4/README.md
Normal file
62
adapters/sweep-1e4/README.md
Normal file
@@ -0,0 +1,62 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
model_name: sweep-1e4
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
licence: license
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for sweep-1e4
|
||||||
|
|
||||||
|
This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B).
|
||||||
|
It has been trained using [TRL](https://github.com/huggingface/trl).
|
||||||
|
|
||||||
|
## Quick start
|
||||||
|
|
||||||
|
```python
|
||||||
|
from transformers import pipeline
|
||||||
|
|
||||||
|
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
||||||
|
generator = pipeline("text-generation", model="None", device="cuda")
|
||||||
|
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
||||||
|
print(output["generated_text"])
|
||||||
|
```
|
||||||
|
|
||||||
|
## Training procedure
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
This model was trained with SFT.
|
||||||
|
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
|
- TRL: 1.8.0
|
||||||
|
- Transformers: 5.13.1
|
||||||
|
- Pytorch: 2.8.0+cu128
|
||||||
|
- Datasets: 5.0.0
|
||||||
|
- Tokenizers: 0.22.2
|
||||||
|
|
||||||
|
## Citations
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
Cite TRL as:
|
||||||
|
|
||||||
|
```bibtex
|
||||||
|
@software{vonwerra2020trl,
|
||||||
|
title = {{TRL: Transformers Reinforcement Learning}},
|
||||||
|
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
||||||
|
license = {Apache-2.0},
|
||||||
|
url = {https://github.com/huggingface/trl},
|
||||||
|
year = {2020}
|
||||||
|
}
|
||||||
|
```
|
||||||
48
adapters/sweep-1e4/adapter_config.json
Normal file
48
adapters/sweep-1e4/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"k_proj",
|
||||||
|
"up_proj",
|
||||||
|
"v_proj",
|
||||||
|
"o_proj",
|
||||||
|
"q_proj",
|
||||||
|
"gate_proj",
|
||||||
|
"down_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
3
adapters/sweep-1e4/adapter_model.safetensors
Normal file
3
adapters/sweep-1e4/adapter_model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:c609c9b881664c2658d1f3dcdc8cd6b6832ca9be8c61a3055f490414f4abcf83
|
||||||
|
size 40422168
|
||||||
89
adapters/sweep-1e4/chat_template.jinja
Normal file
89
adapters/sweep-1e4/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
209
adapters/sweep-1e4/checkpoint-5/README.md
Normal file
209
adapters/sweep-1e4/checkpoint-5/README.md
Normal file
@@ -0,0 +1,209 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for Model ID
|
||||||
|
|
||||||
|
<!-- Provide a quick summary of what the model is/does. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Details
|
||||||
|
|
||||||
|
### Model Description
|
||||||
|
|
||||||
|
<!-- Provide a longer summary of what this model is. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
- **Developed by:** [More Information Needed]
|
||||||
|
- **Funded by [optional]:** [More Information Needed]
|
||||||
|
- **Shared by [optional]:** [More Information Needed]
|
||||||
|
- **Model type:** [More Information Needed]
|
||||||
|
- **Language(s) (NLP):** [More Information Needed]
|
||||||
|
- **License:** [More Information Needed]
|
||||||
|
- **Finetuned from model [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
### Model Sources [optional]
|
||||||
|
|
||||||
|
<!-- Provide the basic links for the model. -->
|
||||||
|
|
||||||
|
- **Repository:** [More Information Needed]
|
||||||
|
- **Paper [optional]:** [More Information Needed]
|
||||||
|
- **Demo [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
## Uses
|
||||||
|
|
||||||
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
||||||
|
|
||||||
|
### Direct Use
|
||||||
|
|
||||||
|
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Downstream Use [optional]
|
||||||
|
|
||||||
|
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Out-of-Scope Use
|
||||||
|
|
||||||
|
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Bias, Risks, and Limitations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Recommendations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
||||||
|
|
||||||
|
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
||||||
|
|
||||||
|
## How to Get Started with the Model
|
||||||
|
|
||||||
|
Use the code below to get started with the model.
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Training Details
|
||||||
|
|
||||||
|
### Training Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Training Procedure
|
||||||
|
|
||||||
|
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
||||||
|
|
||||||
|
#### Preprocessing [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
|
||||||
|
#### Training Hyperparameters
|
||||||
|
|
||||||
|
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
||||||
|
|
||||||
|
#### Speeds, Sizes, Times [optional]
|
||||||
|
|
||||||
|
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Evaluation
|
||||||
|
|
||||||
|
<!-- This section describes the evaluation protocols and provides the results. -->
|
||||||
|
|
||||||
|
### Testing Data, Factors & Metrics
|
||||||
|
|
||||||
|
#### Testing Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card if possible. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Factors
|
||||||
|
|
||||||
|
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Metrics
|
||||||
|
|
||||||
|
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Results
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Summary
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Examination [optional]
|
||||||
|
|
||||||
|
<!-- Relevant interpretability work for the model goes here -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Environmental Impact
|
||||||
|
|
||||||
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
||||||
|
|
||||||
|
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
||||||
|
|
||||||
|
- **Hardware Type:** [More Information Needed]
|
||||||
|
- **Hours used:** [More Information Needed]
|
||||||
|
- **Cloud Provider:** [More Information Needed]
|
||||||
|
- **Compute Region:** [More Information Needed]
|
||||||
|
- **Carbon Emitted:** [More Information Needed]
|
||||||
|
|
||||||
|
## Technical Specifications [optional]
|
||||||
|
|
||||||
|
### Model Architecture and Objective
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Compute Infrastructure
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Hardware
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Software
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Citation [optional]
|
||||||
|
|
||||||
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
||||||
|
|
||||||
|
**BibTeX:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
**APA:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Glossary [optional]
|
||||||
|
|
||||||
|
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## More Information [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Authors [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Contact
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
48
adapters/sweep-1e4/checkpoint-5/adapter_config.json
Normal file
48
adapters/sweep-1e4/checkpoint-5/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"k_proj",
|
||||||
|
"up_proj",
|
||||||
|
"v_proj",
|
||||||
|
"o_proj",
|
||||||
|
"q_proj",
|
||||||
|
"gate_proj",
|
||||||
|
"down_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:c609c9b881664c2658d1f3dcdc8cd6b6832ca9be8c61a3055f490414f4abcf83
|
||||||
|
size 40422168
|
||||||
89
adapters/sweep-1e4/checkpoint-5/chat_template.jinja
Normal file
89
adapters/sweep-1e4/checkpoint-5/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
3
adapters/sweep-1e4/checkpoint-5/optimizer.pt
Normal file
3
adapters/sweep-1e4/checkpoint-5/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:9f483650d01b55513341b32d26f3a5f0568d821f5e92249c0e9c66ba1b9ba864
|
||||||
|
size 81075515
|
||||||
3
adapters/sweep-1e4/checkpoint-5/rng_state.pth
Normal file
3
adapters/sweep-1e4/checkpoint-5/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:812e51d5e3ab4c5b89cb97fe3b9af24d14f1bfc3f1e51a899f18e799a900b918
|
||||||
|
size 14645
|
||||||
3
adapters/sweep-1e4/checkpoint-5/scheduler.pt
Normal file
3
adapters/sweep-1e4/checkpoint-5/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:82ca6510781862cdbebf6aa7638d446745c0e6c20a990f6bd0b5c556992dbd01
|
||||||
|
size 1465
|
||||||
3
adapters/sweep-1e4/checkpoint-5/tokenizer.json
Normal file
3
adapters/sweep-1e4/checkpoint-5/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sweep-1e4/checkpoint-5/tokenizer_config.json
Normal file
30
adapters/sweep-1e4/checkpoint-5/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
117
adapters/sweep-1e4/checkpoint-5/trainer_state.json
Normal file
117
adapters/sweep-1e4/checkpoint-5/trainer_state.json
Normal file
@@ -0,0 +1,117 @@
|
|||||||
|
{
|
||||||
|
"best_global_step": null,
|
||||||
|
"best_metric": null,
|
||||||
|
"best_model_checkpoint": null,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"eval_steps": 2,
|
||||||
|
"global_step": 5,
|
||||||
|
"is_hyper_param_search": false,
|
||||||
|
"is_local_process_zero": true,
|
||||||
|
"is_world_process_zero": true,
|
||||||
|
"log_history": [
|
||||||
|
{
|
||||||
|
"entropy": 0.5484622716903687,
|
||||||
|
"epoch": 0.2,
|
||||||
|
"grad_norm": 4.282809734344482,
|
||||||
|
"learning_rate": 0.0,
|
||||||
|
"loss": 1.4888970851898193,
|
||||||
|
"mean_token_accuracy": 0.7744107842445374,
|
||||||
|
"num_tokens": 11224.0,
|
||||||
|
"step": 1
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.42378363013267517,
|
||||||
|
"epoch": 0.4,
|
||||||
|
"grad_norm": 5.2397661209106445,
|
||||||
|
"learning_rate": 0.0001,
|
||||||
|
"loss": 1.1912707090377808,
|
||||||
|
"mean_token_accuracy": 0.82523512840271,
|
||||||
|
"num_tokens": 22945.0,
|
||||||
|
"step": 2
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4,
|
||||||
|
"eval_entropy": 0.5841028889020284,
|
||||||
|
"eval_loss": 1.3228631019592285,
|
||||||
|
"eval_mean_token_accuracy": 0.7661338647206625,
|
||||||
|
"eval_num_tokens": 22945.0,
|
||||||
|
"eval_runtime": 1.6102,
|
||||||
|
"eval_samples_per_second": 45.956,
|
||||||
|
"eval_steps_per_second": 1.863,
|
||||||
|
"step": 2
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.42866313457489014,
|
||||||
|
"epoch": 0.6,
|
||||||
|
"grad_norm": 3.391592025756836,
|
||||||
|
"learning_rate": 8.535533905932738e-05,
|
||||||
|
"loss": 1.156782865524292,
|
||||||
|
"mean_token_accuracy": 0.8099516034126282,
|
||||||
|
"num_tokens": 33789.0,
|
||||||
|
"step": 3
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7180255055427551,
|
||||||
|
"epoch": 0.8,
|
||||||
|
"grad_norm": 3.040031671524048,
|
||||||
|
"learning_rate": 5e-05,
|
||||||
|
"loss": 1.3655576705932617,
|
||||||
|
"mean_token_accuracy": 0.7426273226737976,
|
||||||
|
"num_tokens": 45381.0,
|
||||||
|
"step": 4
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.8,
|
||||||
|
"eval_entropy": 0.6125266949335734,
|
||||||
|
"eval_loss": 1.1156907081604004,
|
||||||
|
"eval_mean_token_accuracy": 0.7740189035733541,
|
||||||
|
"eval_num_tokens": 45381.0,
|
||||||
|
"eval_runtime": 1.6088,
|
||||||
|
"eval_samples_per_second": 45.997,
|
||||||
|
"eval_steps_per_second": 1.865,
|
||||||
|
"step": 4
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5204336047172546,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"grad_norm": 3.008509397506714,
|
||||||
|
"learning_rate": 1.4644660940672627e-05,
|
||||||
|
"loss": 1.1955260038375854,
|
||||||
|
"mean_token_accuracy": 0.7777777910232544,
|
||||||
|
"num_tokens": 49937.0,
|
||||||
|
"step": 5
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.0,
|
||||||
|
"eval_entropy": 0.6165028810501099,
|
||||||
|
"eval_loss": 1.0987825393676758,
|
||||||
|
"eval_mean_token_accuracy": 0.7757837176322937,
|
||||||
|
"eval_num_tokens": 49937.0,
|
||||||
|
"eval_runtime": 1.6071,
|
||||||
|
"eval_samples_per_second": 46.047,
|
||||||
|
"eval_steps_per_second": 1.867,
|
||||||
|
"step": 5
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"logging_steps": 1,
|
||||||
|
"max_steps": 5,
|
||||||
|
"num_input_tokens_seen": 0,
|
||||||
|
"num_train_epochs": 1,
|
||||||
|
"save_steps": 500,
|
||||||
|
"stateful_callbacks": {
|
||||||
|
"TrainerControl": {
|
||||||
|
"args": {
|
||||||
|
"should_epoch_stop": false,
|
||||||
|
"should_evaluate": false,
|
||||||
|
"should_log": false,
|
||||||
|
"should_save": true,
|
||||||
|
"should_training_stop": true
|
||||||
|
},
|
||||||
|
"attributes": {}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"total_flos": 194982543360000.0,
|
||||||
|
"train_batch_size": 32,
|
||||||
|
"trial_name": null,
|
||||||
|
"trial_params": null
|
||||||
|
}
|
||||||
3
adapters/sweep-1e4/checkpoint-5/training_args.bin
Normal file
3
adapters/sweep-1e4/checkpoint-5/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:7581d827a65c0f2c680c9dce7ea86293760e57899a996736de7dedf39ef9f9d5
|
||||||
|
size 5713
|
||||||
3
adapters/sweep-1e4/tokenizer.json
Normal file
3
adapters/sweep-1e4/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sweep-1e4/tokenizer_config.json
Normal file
30
adapters/sweep-1e4/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
3
adapters/sweep-1e4/training_args.bin
Normal file
3
adapters/sweep-1e4/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:7581d827a65c0f2c680c9dce7ea86293760e57899a996736de7dedf39ef9f9d5
|
||||||
|
size 5713
|
||||||
62
adapters/sweep-2e4/README.md
Normal file
62
adapters/sweep-2e4/README.md
Normal file
@@ -0,0 +1,62 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
model_name: sweep-2e4
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
licence: license
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for sweep-2e4
|
||||||
|
|
||||||
|
This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B).
|
||||||
|
It has been trained using [TRL](https://github.com/huggingface/trl).
|
||||||
|
|
||||||
|
## Quick start
|
||||||
|
|
||||||
|
```python
|
||||||
|
from transformers import pipeline
|
||||||
|
|
||||||
|
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
||||||
|
generator = pipeline("text-generation", model="None", device="cuda")
|
||||||
|
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
||||||
|
print(output["generated_text"])
|
||||||
|
```
|
||||||
|
|
||||||
|
## Training procedure
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
This model was trained with SFT.
|
||||||
|
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
|
- TRL: 1.8.0
|
||||||
|
- Transformers: 5.13.1
|
||||||
|
- Pytorch: 2.8.0+cu128
|
||||||
|
- Datasets: 5.0.0
|
||||||
|
- Tokenizers: 0.22.2
|
||||||
|
|
||||||
|
## Citations
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
Cite TRL as:
|
||||||
|
|
||||||
|
```bibtex
|
||||||
|
@software{vonwerra2020trl,
|
||||||
|
title = {{TRL: Transformers Reinforcement Learning}},
|
||||||
|
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
||||||
|
license = {Apache-2.0},
|
||||||
|
url = {https://github.com/huggingface/trl},
|
||||||
|
year = {2020}
|
||||||
|
}
|
||||||
|
```
|
||||||
48
adapters/sweep-2e4/adapter_config.json
Normal file
48
adapters/sweep-2e4/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"k_proj",
|
||||||
|
"gate_proj",
|
||||||
|
"up_proj",
|
||||||
|
"q_proj",
|
||||||
|
"down_proj",
|
||||||
|
"v_proj",
|
||||||
|
"o_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
3
adapters/sweep-2e4/adapter_model.safetensors
Normal file
3
adapters/sweep-2e4/adapter_model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:b8061ae29549abdc57cbdb1018306f49afed02efef5438e520581f5575cdbff2
|
||||||
|
size 40422168
|
||||||
89
adapters/sweep-2e4/chat_template.jinja
Normal file
89
adapters/sweep-2e4/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
209
adapters/sweep-2e4/checkpoint-5/README.md
Normal file
209
adapters/sweep-2e4/checkpoint-5/README.md
Normal file
@@ -0,0 +1,209 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for Model ID
|
||||||
|
|
||||||
|
<!-- Provide a quick summary of what the model is/does. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Details
|
||||||
|
|
||||||
|
### Model Description
|
||||||
|
|
||||||
|
<!-- Provide a longer summary of what this model is. -->
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
- **Developed by:** [More Information Needed]
|
||||||
|
- **Funded by [optional]:** [More Information Needed]
|
||||||
|
- **Shared by [optional]:** [More Information Needed]
|
||||||
|
- **Model type:** [More Information Needed]
|
||||||
|
- **Language(s) (NLP):** [More Information Needed]
|
||||||
|
- **License:** [More Information Needed]
|
||||||
|
- **Finetuned from model [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
### Model Sources [optional]
|
||||||
|
|
||||||
|
<!-- Provide the basic links for the model. -->
|
||||||
|
|
||||||
|
- **Repository:** [More Information Needed]
|
||||||
|
- **Paper [optional]:** [More Information Needed]
|
||||||
|
- **Demo [optional]:** [More Information Needed]
|
||||||
|
|
||||||
|
## Uses
|
||||||
|
|
||||||
|
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
||||||
|
|
||||||
|
### Direct Use
|
||||||
|
|
||||||
|
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Downstream Use [optional]
|
||||||
|
|
||||||
|
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Out-of-Scope Use
|
||||||
|
|
||||||
|
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Bias, Risks, and Limitations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Recommendations
|
||||||
|
|
||||||
|
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
||||||
|
|
||||||
|
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
||||||
|
|
||||||
|
## How to Get Started with the Model
|
||||||
|
|
||||||
|
Use the code below to get started with the model.
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Training Details
|
||||||
|
|
||||||
|
### Training Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Training Procedure
|
||||||
|
|
||||||
|
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
||||||
|
|
||||||
|
#### Preprocessing [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
|
||||||
|
#### Training Hyperparameters
|
||||||
|
|
||||||
|
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
||||||
|
|
||||||
|
#### Speeds, Sizes, Times [optional]
|
||||||
|
|
||||||
|
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Evaluation
|
||||||
|
|
||||||
|
<!-- This section describes the evaluation protocols and provides the results. -->
|
||||||
|
|
||||||
|
### Testing Data, Factors & Metrics
|
||||||
|
|
||||||
|
#### Testing Data
|
||||||
|
|
||||||
|
<!-- This should link to a Dataset Card if possible. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Factors
|
||||||
|
|
||||||
|
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Metrics
|
||||||
|
|
||||||
|
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Results
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Summary
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
## Model Examination [optional]
|
||||||
|
|
||||||
|
<!-- Relevant interpretability work for the model goes here -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Environmental Impact
|
||||||
|
|
||||||
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
||||||
|
|
||||||
|
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
||||||
|
|
||||||
|
- **Hardware Type:** [More Information Needed]
|
||||||
|
- **Hours used:** [More Information Needed]
|
||||||
|
- **Cloud Provider:** [More Information Needed]
|
||||||
|
- **Compute Region:** [More Information Needed]
|
||||||
|
- **Carbon Emitted:** [More Information Needed]
|
||||||
|
|
||||||
|
## Technical Specifications [optional]
|
||||||
|
|
||||||
|
### Model Architecture and Objective
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
### Compute Infrastructure
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Hardware
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
#### Software
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Citation [optional]
|
||||||
|
|
||||||
|
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
||||||
|
|
||||||
|
**BibTeX:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
**APA:**
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Glossary [optional]
|
||||||
|
|
||||||
|
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## More Information [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Authors [optional]
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
|
||||||
|
## Model Card Contact
|
||||||
|
|
||||||
|
[More Information Needed]
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
48
adapters/sweep-2e4/checkpoint-5/adapter_config.json
Normal file
48
adapters/sweep-2e4/checkpoint-5/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"k_proj",
|
||||||
|
"gate_proj",
|
||||||
|
"up_proj",
|
||||||
|
"q_proj",
|
||||||
|
"down_proj",
|
||||||
|
"v_proj",
|
||||||
|
"o_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:b8061ae29549abdc57cbdb1018306f49afed02efef5438e520581f5575cdbff2
|
||||||
|
size 40422168
|
||||||
89
adapters/sweep-2e4/checkpoint-5/chat_template.jinja
Normal file
89
adapters/sweep-2e4/checkpoint-5/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
3
adapters/sweep-2e4/checkpoint-5/optimizer.pt
Normal file
3
adapters/sweep-2e4/checkpoint-5/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:c9c95f873adfc1d54ef8cb58eea7582d6c7b8c5db6239ad7112d2b61d7e88745
|
||||||
|
size 81075515
|
||||||
3
adapters/sweep-2e4/checkpoint-5/rng_state.pth
Normal file
3
adapters/sweep-2e4/checkpoint-5/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:812e51d5e3ab4c5b89cb97fe3b9af24d14f1bfc3f1e51a899f18e799a900b918
|
||||||
|
size 14645
|
||||||
3
adapters/sweep-2e4/checkpoint-5/scheduler.pt
Normal file
3
adapters/sweep-2e4/checkpoint-5/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:86fb17cdadef78d9db4408e9cb9fca613afe30dca27e3c6738659beae16b7c9c
|
||||||
|
size 1465
|
||||||
3
adapters/sweep-2e4/checkpoint-5/tokenizer.json
Normal file
3
adapters/sweep-2e4/checkpoint-5/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sweep-2e4/checkpoint-5/tokenizer_config.json
Normal file
30
adapters/sweep-2e4/checkpoint-5/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
117
adapters/sweep-2e4/checkpoint-5/trainer_state.json
Normal file
117
adapters/sweep-2e4/checkpoint-5/trainer_state.json
Normal file
@@ -0,0 +1,117 @@
|
|||||||
|
{
|
||||||
|
"best_global_step": null,
|
||||||
|
"best_metric": null,
|
||||||
|
"best_model_checkpoint": null,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"eval_steps": 2,
|
||||||
|
"global_step": 5,
|
||||||
|
"is_hyper_param_search": false,
|
||||||
|
"is_local_process_zero": true,
|
||||||
|
"is_world_process_zero": true,
|
||||||
|
"log_history": [
|
||||||
|
{
|
||||||
|
"entropy": 0.5484622716903687,
|
||||||
|
"epoch": 0.2,
|
||||||
|
"grad_norm": 4.228063106536865,
|
||||||
|
"learning_rate": 0.0,
|
||||||
|
"loss": 1.4888970851898193,
|
||||||
|
"mean_token_accuracy": 0.7744107842445374,
|
||||||
|
"num_tokens": 11224.0,
|
||||||
|
"step": 1
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.42378363013267517,
|
||||||
|
"epoch": 0.4,
|
||||||
|
"grad_norm": 5.219785690307617,
|
||||||
|
"learning_rate": 0.0002,
|
||||||
|
"loss": 1.1912707090377808,
|
||||||
|
"mean_token_accuracy": 0.82523512840271,
|
||||||
|
"num_tokens": 22945.0,
|
||||||
|
"step": 2
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.4,
|
||||||
|
"eval_entropy": 0.5992160042126974,
|
||||||
|
"eval_loss": 1.226914644241333,
|
||||||
|
"eval_mean_token_accuracy": 0.757758637269338,
|
||||||
|
"eval_num_tokens": 22945.0,
|
||||||
|
"eval_runtime": 1.6104,
|
||||||
|
"eval_samples_per_second": 45.951,
|
||||||
|
"eval_steps_per_second": 1.863,
|
||||||
|
"step": 2
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.4446622431278229,
|
||||||
|
"epoch": 0.6,
|
||||||
|
"grad_norm": 3.786647319793701,
|
||||||
|
"learning_rate": 0.00017071067811865476,
|
||||||
|
"loss": 1.0181313753128052,
|
||||||
|
"mean_token_accuracy": 0.8030408024787903,
|
||||||
|
"num_tokens": 33789.0,
|
||||||
|
"step": 3
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.7599555850028992,
|
||||||
|
"epoch": 0.8,
|
||||||
|
"grad_norm": 1.5579856634140015,
|
||||||
|
"learning_rate": 0.0001,
|
||||||
|
"loss": 1.2033650875091553,
|
||||||
|
"mean_token_accuracy": 0.7513405084609985,
|
||||||
|
"num_tokens": 45381.0,
|
||||||
|
"step": 4
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 0.8,
|
||||||
|
"eval_entropy": 0.6641819675763448,
|
||||||
|
"eval_loss": 1.0074481964111328,
|
||||||
|
"eval_mean_token_accuracy": 0.7918139894803365,
|
||||||
|
"eval_num_tokens": 45381.0,
|
||||||
|
"eval_runtime": 1.6119,
|
||||||
|
"eval_samples_per_second": 45.908,
|
||||||
|
"eval_steps_per_second": 1.861,
|
||||||
|
"step": 4
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"entropy": 0.5738851428031921,
|
||||||
|
"epoch": 1.0,
|
||||||
|
"grad_norm": 1.8575931787490845,
|
||||||
|
"learning_rate": 2.9289321881345254e-05,
|
||||||
|
"loss": 1.0246576070785522,
|
||||||
|
"mean_token_accuracy": 0.8115941882133484,
|
||||||
|
"num_tokens": 49937.0,
|
||||||
|
"step": 5
|
||||||
|
},
|
||||||
|
{
|
||||||
|
"epoch": 1.0,
|
||||||
|
"eval_entropy": 0.6727284789085388,
|
||||||
|
"eval_loss": 0.999213457107544,
|
||||||
|
"eval_mean_token_accuracy": 0.7922166585922241,
|
||||||
|
"eval_num_tokens": 49937.0,
|
||||||
|
"eval_runtime": 1.612,
|
||||||
|
"eval_samples_per_second": 45.907,
|
||||||
|
"eval_steps_per_second": 1.861,
|
||||||
|
"step": 5
|
||||||
|
}
|
||||||
|
],
|
||||||
|
"logging_steps": 1,
|
||||||
|
"max_steps": 5,
|
||||||
|
"num_input_tokens_seen": 0,
|
||||||
|
"num_train_epochs": 1,
|
||||||
|
"save_steps": 500,
|
||||||
|
"stateful_callbacks": {
|
||||||
|
"TrainerControl": {
|
||||||
|
"args": {
|
||||||
|
"should_epoch_stop": false,
|
||||||
|
"should_evaluate": false,
|
||||||
|
"should_log": false,
|
||||||
|
"should_save": true,
|
||||||
|
"should_training_stop": true
|
||||||
|
},
|
||||||
|
"attributes": {}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
"total_flos": 194982543360000.0,
|
||||||
|
"train_batch_size": 32,
|
||||||
|
"trial_name": null,
|
||||||
|
"trial_params": null
|
||||||
|
}
|
||||||
3
adapters/sweep-2e4/checkpoint-5/training_args.bin
Normal file
3
adapters/sweep-2e4/checkpoint-5/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:d5eac8dce8bbd7c707b47c1e3da6573bc33e576388a94b661710aaf52b8e371d
|
||||||
|
size 5713
|
||||||
3
adapters/sweep-2e4/tokenizer.json
Normal file
3
adapters/sweep-2e4/tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506
|
||||||
|
size 11422650
|
||||||
30
adapters/sweep-2e4/tokenizer_config.json
Normal file
30
adapters/sweep-2e4/tokenizer_config.json
Normal file
@@ -0,0 +1,30 @@
|
|||||||
|
{
|
||||||
|
"add_prefix_space": false,
|
||||||
|
"backend": "tokenizers",
|
||||||
|
"bos_token": null,
|
||||||
|
"clean_up_tokenization_spaces": false,
|
||||||
|
"eos_token": "<|im_end|>",
|
||||||
|
"errors": "replace",
|
||||||
|
"extra_special_tokens": [
|
||||||
|
"<|im_start|>",
|
||||||
|
"<|im_end|>",
|
||||||
|
"<|object_ref_start|>",
|
||||||
|
"<|object_ref_end|>",
|
||||||
|
"<|box_start|>",
|
||||||
|
"<|box_end|>",
|
||||||
|
"<|quad_start|>",
|
||||||
|
"<|quad_end|>",
|
||||||
|
"<|vision_start|>",
|
||||||
|
"<|vision_end|>",
|
||||||
|
"<|vision_pad|>",
|
||||||
|
"<|image_pad|>",
|
||||||
|
"<|video_pad|>"
|
||||||
|
],
|
||||||
|
"is_local": false,
|
||||||
|
"local_files_only": false,
|
||||||
|
"model_max_length": 131072,
|
||||||
|
"pad_token": "<|endoftext|>",
|
||||||
|
"split_special_tokens": false,
|
||||||
|
"tokenizer_class": "Qwen2Tokenizer",
|
||||||
|
"unk_token": null
|
||||||
|
}
|
||||||
3
adapters/sweep-2e4/training_args.bin
Normal file
3
adapters/sweep-2e4/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:d5eac8dce8bbd7c707b47c1e3da6573bc33e576388a94b661710aaf52b8e371d
|
||||||
|
size 5713
|
||||||
62
adapters/sweep-3e4/README.md
Normal file
62
adapters/sweep-3e4/README.md
Normal file
@@ -0,0 +1,62 @@
|
|||||||
|
---
|
||||||
|
base_model: Qwen/Qwen3-0.6B
|
||||||
|
library_name: peft
|
||||||
|
model_name: sweep-3e4
|
||||||
|
tags:
|
||||||
|
- base_model:adapter:Qwen/Qwen3-0.6B
|
||||||
|
- lora
|
||||||
|
- sft
|
||||||
|
- transformers
|
||||||
|
- trl
|
||||||
|
licence: license
|
||||||
|
pipeline_tag: text-generation
|
||||||
|
---
|
||||||
|
|
||||||
|
# Model Card for sweep-3e4
|
||||||
|
|
||||||
|
This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B).
|
||||||
|
It has been trained using [TRL](https://github.com/huggingface/trl).
|
||||||
|
|
||||||
|
## Quick start
|
||||||
|
|
||||||
|
```python
|
||||||
|
from transformers import pipeline
|
||||||
|
|
||||||
|
question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
|
||||||
|
generator = pipeline("text-generation", model="None", device="cuda")
|
||||||
|
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
|
||||||
|
print(output["generated_text"])
|
||||||
|
```
|
||||||
|
|
||||||
|
## Training procedure
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
This model was trained with SFT.
|
||||||
|
|
||||||
|
### Framework versions
|
||||||
|
|
||||||
|
- PEFT 0.19.1
|
||||||
|
- TRL: 1.8.0
|
||||||
|
- Transformers: 5.13.1
|
||||||
|
- Pytorch: 2.8.0+cu128
|
||||||
|
- Datasets: 5.0.0
|
||||||
|
- Tokenizers: 0.22.2
|
||||||
|
|
||||||
|
## Citations
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
Cite TRL as:
|
||||||
|
|
||||||
|
```bibtex
|
||||||
|
@software{vonwerra2020trl,
|
||||||
|
title = {{TRL: Transformers Reinforcement Learning}},
|
||||||
|
author = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
|
||||||
|
license = {Apache-2.0},
|
||||||
|
url = {https://github.com/huggingface/trl},
|
||||||
|
year = {2020}
|
||||||
|
}
|
||||||
|
```
|
||||||
48
adapters/sweep-3e4/adapter_config.json
Normal file
48
adapters/sweep-3e4/adapter_config.json
Normal file
@@ -0,0 +1,48 @@
|
|||||||
|
{
|
||||||
|
"alora_invocation_tokens": null,
|
||||||
|
"alpha_pattern": {},
|
||||||
|
"arrow_config": null,
|
||||||
|
"auto_mapping": null,
|
||||||
|
"base_model_name_or_path": "Qwen/Qwen3-0.6B",
|
||||||
|
"bias": "none",
|
||||||
|
"corda_config": null,
|
||||||
|
"ensure_weight_tying": false,
|
||||||
|
"eva_config": null,
|
||||||
|
"exclude_modules": null,
|
||||||
|
"fan_in_fan_out": false,
|
||||||
|
"inference_mode": true,
|
||||||
|
"init_lora_weights": true,
|
||||||
|
"layer_replication": null,
|
||||||
|
"layers_pattern": null,
|
||||||
|
"layers_to_transform": null,
|
||||||
|
"loftq_config": {},
|
||||||
|
"lora_alpha": 32,
|
||||||
|
"lora_bias": false,
|
||||||
|
"lora_dropout": 0.05,
|
||||||
|
"lora_ga_config": null,
|
||||||
|
"megatron_config": null,
|
||||||
|
"megatron_core": "megatron.core",
|
||||||
|
"modules_to_save": null,
|
||||||
|
"peft_type": "LORA",
|
||||||
|
"peft_version": "0.19.1",
|
||||||
|
"qalora_group_size": 16,
|
||||||
|
"r": 16,
|
||||||
|
"rank_pattern": {},
|
||||||
|
"revision": null,
|
||||||
|
"target_modules": [
|
||||||
|
"down_proj",
|
||||||
|
"k_proj",
|
||||||
|
"v_proj",
|
||||||
|
"up_proj",
|
||||||
|
"gate_proj",
|
||||||
|
"o_proj",
|
||||||
|
"q_proj"
|
||||||
|
],
|
||||||
|
"target_parameters": null,
|
||||||
|
"task_type": "CAUSAL_LM",
|
||||||
|
"trainable_token_indices": null,
|
||||||
|
"use_bdlora": null,
|
||||||
|
"use_dora": false,
|
||||||
|
"use_qalora": false,
|
||||||
|
"use_rslora": false
|
||||||
|
}
|
||||||
3
adapters/sweep-3e4/adapter_model.safetensors
Normal file
3
adapters/sweep-3e4/adapter_model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
|||||||
|
version https://git-lfs.github.com/spec/v1
|
||||||
|
oid sha256:9ed480425b0dde1dc5cc5bbde4ddfc8d49b8b9cd73c6916710c15334be1a2b11
|
||||||
|
size 40422168
|
||||||
89
adapters/sweep-3e4/chat_template.jinja
Normal file
89
adapters/sweep-3e4/chat_template.jinja
Normal file
@@ -0,0 +1,89 @@
|
|||||||
|
{%- if tools %}
|
||||||
|
{{- '<|im_start|>system\n' }}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- messages[0].content + '\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
||||||
|
{%- for tool in tools %}
|
||||||
|
{{- "\n" }}
|
||||||
|
{{- tool | tojson }}
|
||||||
|
{%- endfor %}
|
||||||
|
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
||||||
|
{%- else %}
|
||||||
|
{%- if messages[0].role == 'system' %}
|
||||||
|
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
|
||||||
|
{%- for message in messages[::-1] %}
|
||||||
|
{%- set index = (messages|length - 1) - loop.index0 %}
|
||||||
|
{%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
|
||||||
|
{%- set ns.multi_step_tool = false %}
|
||||||
|
{%- set ns.last_query_index = index %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- for message in messages %}
|
||||||
|
{%- if message.content is string %}
|
||||||
|
{%- set content = message.content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- set content = '' %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
||||||
|
{%- elif message.role == "assistant" %}
|
||||||
|
{%- set reasoning_content = '' %}
|
||||||
|
{%- if message.reasoning_content is string %}
|
||||||
|
{%- set reasoning_content = message.reasoning_content %}
|
||||||
|
{%- else %}
|
||||||
|
{%- if '</think>' in content %}
|
||||||
|
{%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
|
||||||
|
{%- set content = content.split('</think>')[-1].lstrip('\n') %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if loop.index0 > ns.last_query_index %}
|
||||||
|
{%- if loop.last or (not loop.last and reasoning_content) %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- else %}
|
||||||
|
{{- '<|im_start|>' + message.role + '\n' + content }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if message.tool_calls %}
|
||||||
|
{%- for tool_call in message.tool_calls %}
|
||||||
|
{%- if (loop.first and content) or (not loop.first) %}
|
||||||
|
{{- '\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- if tool_call.function %}
|
||||||
|
{%- set tool_call = tool_call.function %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<tool_call>\n{"name": "' }}
|
||||||
|
{{- tool_call.name }}
|
||||||
|
{{- '", "arguments": ' }}
|
||||||
|
{%- if tool_call.arguments is string %}
|
||||||
|
{{- tool_call.arguments }}
|
||||||
|
{%- else %}
|
||||||
|
{{- tool_call.arguments | tojson }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '}\n</tool_call>' }}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- elif message.role == "tool" %}
|
||||||
|
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
||||||
|
{{- '<|im_start|>user' }}
|
||||||
|
{%- endif %}
|
||||||
|
{{- '\n<tool_response>\n' }}
|
||||||
|
{{- content }}
|
||||||
|
{{- '\n</tool_response>' }}
|
||||||
|
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
||||||
|
{{- '<|im_end|>\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endfor %}
|
||||||
|
{%- if add_generation_prompt %}
|
||||||
|
{{- '<|im_start|>assistant\n' }}
|
||||||
|
{%- if enable_thinking is defined and enable_thinking is false %}
|
||||||
|
{{- '<think>\n\n</think>\n\n' }}
|
||||||
|
{%- endif %}
|
||||||
|
{%- endif %}
|
||||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user