初始化项目,由ModelHub XC社区提供模型
Model: theophilusowiti/Caracal_GPT Source: Original Platform
This commit is contained in:
33
checkpoints/checkpoint-86520/config.json
Normal file
33
checkpoints/checkpoint-86520/config.json
Normal file
@@ -0,0 +1,33 @@
|
||||
{
|
||||
"_name_or_path": "lelapa/InkubaLM-0.4B",
|
||||
"architectures": [
|
||||
"LlamaForCausalLM"
|
||||
],
|
||||
"attention_bias": false,
|
||||
"attention_dropout": 0.0,
|
||||
"auto_map": {
|
||||
"AutoModelForCausalLM": "lelapa/InkubaLM-0.4B--vulavulaslm.VulavulaLlamaForCausalLM"
|
||||
},
|
||||
"bos_token_id": 1,
|
||||
"eos_token_id": 2,
|
||||
"head_dim": 64,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 5632,
|
||||
"max_position_embeddings": 2048,
|
||||
"mlp_bias": false,
|
||||
"model_type": "llama",
|
||||
"num_attention_heads": 32,
|
||||
"num_hidden_layers": 8,
|
||||
"num_key_value_heads": 32,
|
||||
"pretraining_tp": 1,
|
||||
"rms_norm_eps": 1e-05,
|
||||
"rope_scaling": null,
|
||||
"rope_theta": 10000.0,
|
||||
"tie_word_embeddings": false,
|
||||
"torch_dtype": "bfloat16",
|
||||
"transformers_version": "4.45.2",
|
||||
"use_cache": true,
|
||||
"vocab_size": 61788
|
||||
}
|
||||
6
checkpoints/checkpoint-86520/generation_config.json
Normal file
6
checkpoints/checkpoint-86520/generation_config.json
Normal file
@@ -0,0 +1,6 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"bos_token_id": 1,
|
||||
"eos_token_id": 2,
|
||||
"transformers_version": "4.45.2"
|
||||
}
|
||||
3
checkpoints/checkpoint-86520/model.safetensors
Normal file
3
checkpoints/checkpoint-86520/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b203873e6abe1bdf6330c91fcc88a909e14096edb4c38aa1a2d03d54422a9480
|
||||
size 1328329064
|
||||
3
checkpoints/checkpoint-86520/optimizer.pt
Normal file
3
checkpoints/checkpoint-86520/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:d168f0101b7ec3e5a38c3cbb323364e57dbe77708164592ae936750a35797315
|
||||
size 2656706059
|
||||
3
checkpoints/checkpoint-86520/rng_state.pth
Normal file
3
checkpoints/checkpoint-86520/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:f016673068a962ff8dbca4327fe70d8b75e3232792c35d2d5d93b4234a6d74e6
|
||||
size 14645
|
||||
3
checkpoints/checkpoint-86520/scheduler.pt
Normal file
3
checkpoints/checkpoint-86520/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:70429f4312c0ff5f72a6d39fe11d46a648b58aba6614383cbbd7537c2ac99e19
|
||||
size 1465
|
||||
24
checkpoints/checkpoint-86520/special_tokens_map.json
Normal file
24
checkpoints/checkpoint-86520/special_tokens_map.json
Normal file
@@ -0,0 +1,24 @@
|
||||
{
|
||||
"bos_token": {
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"eos_token": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"pad_token": "</s>",
|
||||
"unk_token": {
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
}
|
||||
}
|
||||
608412
checkpoints/checkpoint-86520/tokenizer.json
Normal file
608412
checkpoints/checkpoint-86520/tokenizer.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoints/checkpoint-86520/tokenizer.model
Normal file
3
checkpoints/checkpoint-86520/tokenizer.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:1c41fcc6d44fcc4e8269e41dffe0123687baf800bd95a9c8b5d48abd9cb8971b
|
||||
size 991189
|
||||
42
checkpoints/checkpoint-86520/tokenizer_config.json
Normal file
42
checkpoints/checkpoint-86520/tokenizer_config.json
Normal file
@@ -0,0 +1,42 @@
|
||||
{
|
||||
"add_bos_token": true,
|
||||
"add_eos_token": false,
|
||||
"add_prefix_space": true,
|
||||
"added_tokens_decoder": {
|
||||
"0": {
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"1": {
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"2": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
}
|
||||
},
|
||||
"bos_token": "<s>",
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "</s>",
|
||||
"legacy": true,
|
||||
"model_max_length": 1000000000000000019884624838656,
|
||||
"pad_token": "</s>",
|
||||
"sp_model_kwargs": {},
|
||||
"spaces_between_special_tokens": false,
|
||||
"tokenizer_class": "LlamaTokenizer",
|
||||
"unk_token": "<unk>",
|
||||
"use_default_system_prompt": false
|
||||
}
|
||||
12143
checkpoints/checkpoint-86520/trainer_state.json
Normal file
12143
checkpoints/checkpoint-86520/trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoints/checkpoint-86520/training_args.bin
Normal file
3
checkpoints/checkpoint-86520/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:d07c893cfd81696e1ad219e8dba8be536a15aa91ea4b1b878d196fb4d4ca8088
|
||||
size 5713
|
||||
Reference in New Issue
Block a user