Upload folder using huggingface_hub
Browse files- config.json +29 -0
- model.safetensors +3 -0
- special_tokens_map.json +25 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +40 -0
- trainer_state.json +617 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "princeton-nlp/Sheared-LLaMA-1.3B",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaBiModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 2048,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 5504,
|
| 14 |
+
"max_position_embeddings": 4096,
|
| 15 |
+
"model_type": "llama",
|
| 16 |
+
"num_attention_heads": 16,
|
| 17 |
+
"num_hidden_layers": 24,
|
| 18 |
+
"num_key_value_heads": 16,
|
| 19 |
+
"pad_token_id": 0,
|
| 20 |
+
"pretraining_tp": 1,
|
| 21 |
+
"rms_norm_eps": 1e-05,
|
| 22 |
+
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 10000.0,
|
| 24 |
+
"tie_word_embeddings": false,
|
| 25 |
+
"torch_dtype": "bfloat16",
|
| 26 |
+
"transformers_version": "4.40.2",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"vocab_size": 32000
|
| 29 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06b5579fd0d157162fbf371573b4a415c23bb5eec858b575220112927e9ada76
|
| 3 |
+
size 2559798552
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"mask_token": "_",
|
| 17 |
+
"pad_token": "</s>",
|
| 18 |
+
"unk_token": {
|
| 19 |
+
"content": "<unk>",
|
| 20 |
+
"lstrip": false,
|
| 21 |
+
"normalized": true,
|
| 22 |
+
"rstrip": false,
|
| 23 |
+
"single_word": false
|
| 24 |
+
}
|
| 25 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<unk>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": true,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": true,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
}
|
| 29 |
+
},
|
| 30 |
+
"bos_token": "<s>",
|
| 31 |
+
"clean_up_tokenization_spaces": false,
|
| 32 |
+
"eos_token": "</s>",
|
| 33 |
+
"mask_token": "_",
|
| 34 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 35 |
+
"pad_token": "</s>",
|
| 36 |
+
"sp_model_kwargs": {},
|
| 37 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 38 |
+
"unk_token": "<unk>",
|
| 39 |
+
"use_default_system_prompt": false
|
| 40 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,617 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.32923775694404,
|
| 5 |
+
"eval_steps": 10000,
|
| 6 |
+
"global_step": 40000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0291154719618005,
|
| 13 |
+
"grad_norm": 5.6875,
|
| 14 |
+
"learning_rate": 4.9514742133969994e-05,
|
| 15 |
+
"loss": 2.4005,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.058230943923601,
|
| 20 |
+
"grad_norm": 5.125,
|
| 21 |
+
"learning_rate": 4.9029484267939986e-05,
|
| 22 |
+
"loss": 1.8762,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0873464158854015,
|
| 27 |
+
"grad_norm": 5.21875,
|
| 28 |
+
"learning_rate": 4.854422640190998e-05,
|
| 29 |
+
"loss": 1.7508,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.116461887847202,
|
| 34 |
+
"grad_norm": 5.125,
|
| 35 |
+
"learning_rate": 4.805896853587997e-05,
|
| 36 |
+
"loss": 1.6692,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1455773598090025,
|
| 41 |
+
"grad_norm": 5.03125,
|
| 42 |
+
"learning_rate": 4.757371066984996e-05,
|
| 43 |
+
"loss": 1.6309,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.174692831770803,
|
| 48 |
+
"grad_norm": 4.6875,
|
| 49 |
+
"learning_rate": 4.708845280381995e-05,
|
| 50 |
+
"loss": 1.5878,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2038083037326035,
|
| 55 |
+
"grad_norm": 4.90625,
|
| 56 |
+
"learning_rate": 4.6603194937789944e-05,
|
| 57 |
+
"loss": 1.5593,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.232923775694404,
|
| 62 |
+
"grad_norm": 4.65625,
|
| 63 |
+
"learning_rate": 4.6117937071759936e-05,
|
| 64 |
+
"loss": 1.5287,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2620392476562045,
|
| 69 |
+
"grad_norm": 5.53125,
|
| 70 |
+
"learning_rate": 4.563267920572993e-05,
|
| 71 |
+
"loss": 1.5189,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.291154719618005,
|
| 76 |
+
"grad_norm": 4.53125,
|
| 77 |
+
"learning_rate": 4.514742133969992e-05,
|
| 78 |
+
"loss": 1.5018,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3202701915798055,
|
| 83 |
+
"grad_norm": 4.8125,
|
| 84 |
+
"learning_rate": 4.466216347366991e-05,
|
| 85 |
+
"loss": 1.4784,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.349385663541606,
|
| 90 |
+
"grad_norm": 4.6875,
|
| 91 |
+
"learning_rate": 4.41769056076399e-05,
|
| 92 |
+
"loss": 1.4703,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3785011355034065,
|
| 97 |
+
"grad_norm": 4.5,
|
| 98 |
+
"learning_rate": 4.3691647741609894e-05,
|
| 99 |
+
"loss": 1.449,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.407616607465207,
|
| 104 |
+
"grad_norm": 4.71875,
|
| 105 |
+
"learning_rate": 4.3206389875579886e-05,
|
| 106 |
+
"loss": 1.4496,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.43673207942700754,
|
| 111 |
+
"grad_norm": 4.65625,
|
| 112 |
+
"learning_rate": 4.272113200954988e-05,
|
| 113 |
+
"loss": 1.4421,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.465847551388808,
|
| 118 |
+
"grad_norm": 4.90625,
|
| 119 |
+
"learning_rate": 4.223587414351987e-05,
|
| 120 |
+
"loss": 1.4305,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.4949630233506085,
|
| 125 |
+
"grad_norm": 4.25,
|
| 126 |
+
"learning_rate": 4.175061627748986e-05,
|
| 127 |
+
"loss": 1.4251,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.524078495312409,
|
| 132 |
+
"grad_norm": 4.53125,
|
| 133 |
+
"learning_rate": 4.126535841145985e-05,
|
| 134 |
+
"loss": 1.4116,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.5531939672742096,
|
| 139 |
+
"grad_norm": 4.375,
|
| 140 |
+
"learning_rate": 4.0780100545429844e-05,
|
| 141 |
+
"loss": 1.4124,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.58230943923601,
|
| 146 |
+
"grad_norm": 5.40625,
|
| 147 |
+
"learning_rate": 4.0294842679399835e-05,
|
| 148 |
+
"loss": 1.4035,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.58230943923601,
|
| 153 |
+
"eval_accuracy": 0.7037869026306487,
|
| 154 |
+
"eval_loss": 1.3742600679397583,
|
| 155 |
+
"eval_runtime": 6.4605,
|
| 156 |
+
"eval_samples_per_second": 89.931,
|
| 157 |
+
"eval_steps_per_second": 2.941,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.6114249111978105,
|
| 162 |
+
"grad_norm": 4.40625,
|
| 163 |
+
"learning_rate": 3.980958481336983e-05,
|
| 164 |
+
"loss": 1.4042,
|
| 165 |
+
"step": 10500
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.640540383159611,
|
| 169 |
+
"grad_norm": 4.96875,
|
| 170 |
+
"learning_rate": 3.932432694733982e-05,
|
| 171 |
+
"loss": 1.3966,
|
| 172 |
+
"step": 11000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.6696558551214116,
|
| 176 |
+
"grad_norm": 4.5625,
|
| 177 |
+
"learning_rate": 3.883906908130981e-05,
|
| 178 |
+
"loss": 1.3887,
|
| 179 |
+
"step": 11500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.698771327083212,
|
| 183 |
+
"grad_norm": 5.0,
|
| 184 |
+
"learning_rate": 3.83538112152798e-05,
|
| 185 |
+
"loss": 1.3843,
|
| 186 |
+
"step": 12000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.7278867990450125,
|
| 190 |
+
"grad_norm": 4.65625,
|
| 191 |
+
"learning_rate": 3.7868553349249794e-05,
|
| 192 |
+
"loss": 1.3783,
|
| 193 |
+
"step": 12500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.757002271006813,
|
| 197 |
+
"grad_norm": 4.6875,
|
| 198 |
+
"learning_rate": 3.7383295483219785e-05,
|
| 199 |
+
"loss": 1.375,
|
| 200 |
+
"step": 13000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.7861177429686135,
|
| 204 |
+
"grad_norm": 4.46875,
|
| 205 |
+
"learning_rate": 3.689803761718978e-05,
|
| 206 |
+
"loss": 1.3756,
|
| 207 |
+
"step": 13500
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.815233214930414,
|
| 211 |
+
"grad_norm": 4.65625,
|
| 212 |
+
"learning_rate": 3.641277975115977e-05,
|
| 213 |
+
"loss": 1.3636,
|
| 214 |
+
"step": 14000
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.8443486868922145,
|
| 218 |
+
"grad_norm": 4.5625,
|
| 219 |
+
"learning_rate": 3.592752188512976e-05,
|
| 220 |
+
"loss": 1.3688,
|
| 221 |
+
"step": 14500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.8734641588540151,
|
| 225 |
+
"grad_norm": 4.59375,
|
| 226 |
+
"learning_rate": 3.544226401909975e-05,
|
| 227 |
+
"loss": 1.3652,
|
| 228 |
+
"step": 15000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.9025796308158155,
|
| 232 |
+
"grad_norm": 4.90625,
|
| 233 |
+
"learning_rate": 3.4957006153069744e-05,
|
| 234 |
+
"loss": 1.3611,
|
| 235 |
+
"step": 15500
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.931695102777616,
|
| 239 |
+
"grad_norm": 5.09375,
|
| 240 |
+
"learning_rate": 3.4471748287039735e-05,
|
| 241 |
+
"loss": 1.352,
|
| 242 |
+
"step": 16000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.9608105747394166,
|
| 246 |
+
"grad_norm": 4.625,
|
| 247 |
+
"learning_rate": 3.398649042100973e-05,
|
| 248 |
+
"loss": 1.3562,
|
| 249 |
+
"step": 16500
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.989926046701217,
|
| 253 |
+
"grad_norm": 4.84375,
|
| 254 |
+
"learning_rate": 3.350123255497972e-05,
|
| 255 |
+
"loss": 1.3491,
|
| 256 |
+
"step": 17000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.0190415186630175,
|
| 260 |
+
"grad_norm": 5.15625,
|
| 261 |
+
"learning_rate": 3.301597468894971e-05,
|
| 262 |
+
"loss": 1.3448,
|
| 263 |
+
"step": 17500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.048156990624818,
|
| 267 |
+
"grad_norm": 4.625,
|
| 268 |
+
"learning_rate": 3.25307168229197e-05,
|
| 269 |
+
"loss": 1.344,
|
| 270 |
+
"step": 18000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.0772724625866186,
|
| 274 |
+
"grad_norm": 4.75,
|
| 275 |
+
"learning_rate": 3.204545895688969e-05,
|
| 276 |
+
"loss": 1.3381,
|
| 277 |
+
"step": 18500
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 1.106387934548419,
|
| 281 |
+
"grad_norm": 4.8125,
|
| 282 |
+
"learning_rate": 3.1560201090859685e-05,
|
| 283 |
+
"loss": 1.34,
|
| 284 |
+
"step": 19000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.1355034065102196,
|
| 288 |
+
"grad_norm": 5.375,
|
| 289 |
+
"learning_rate": 3.107494322482968e-05,
|
| 290 |
+
"loss": 1.3352,
|
| 291 |
+
"step": 19500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 1.16461887847202,
|
| 295 |
+
"grad_norm": 5.71875,
|
| 296 |
+
"learning_rate": 3.058968535879967e-05,
|
| 297 |
+
"loss": 1.3374,
|
| 298 |
+
"step": 20000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.16461887847202,
|
| 302 |
+
"eval_accuracy": 0.7118764765058947,
|
| 303 |
+
"eval_loss": 1.3248625993728638,
|
| 304 |
+
"eval_runtime": 6.4569,
|
| 305 |
+
"eval_samples_per_second": 89.981,
|
| 306 |
+
"eval_steps_per_second": 2.943,
|
| 307 |
+
"step": 20000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.1937343504338205,
|
| 311 |
+
"grad_norm": 4.5,
|
| 312 |
+
"learning_rate": 3.010442749276966e-05,
|
| 313 |
+
"loss": 1.3314,
|
| 314 |
+
"step": 20500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.222849822395621,
|
| 318 |
+
"grad_norm": 5.09375,
|
| 319 |
+
"learning_rate": 2.9619169626739652e-05,
|
| 320 |
+
"loss": 1.3296,
|
| 321 |
+
"step": 21000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 1.2519652943574215,
|
| 325 |
+
"grad_norm": 4.40625,
|
| 326 |
+
"learning_rate": 2.9133911760709644e-05,
|
| 327 |
+
"loss": 1.3331,
|
| 328 |
+
"step": 21500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.281080766319222,
|
| 332 |
+
"grad_norm": 4.75,
|
| 333 |
+
"learning_rate": 2.8648653894679632e-05,
|
| 334 |
+
"loss": 1.332,
|
| 335 |
+
"step": 22000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 1.3101962382810226,
|
| 339 |
+
"grad_norm": 5.09375,
|
| 340 |
+
"learning_rate": 2.8163396028649623e-05,
|
| 341 |
+
"loss": 1.3291,
|
| 342 |
+
"step": 22500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.3393117102428231,
|
| 346 |
+
"grad_norm": 4.53125,
|
| 347 |
+
"learning_rate": 2.7678138162619615e-05,
|
| 348 |
+
"loss": 1.3301,
|
| 349 |
+
"step": 23000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 1.3684271822046234,
|
| 353 |
+
"grad_norm": 4.8125,
|
| 354 |
+
"learning_rate": 2.7192880296589607e-05,
|
| 355 |
+
"loss": 1.3266,
|
| 356 |
+
"step": 23500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 1.397542654166424,
|
| 360 |
+
"grad_norm": 5.0,
|
| 361 |
+
"learning_rate": 2.67076224305596e-05,
|
| 362 |
+
"loss": 1.3178,
|
| 363 |
+
"step": 24000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.4266581261282245,
|
| 367 |
+
"grad_norm": 5.15625,
|
| 368 |
+
"learning_rate": 2.6222364564529593e-05,
|
| 369 |
+
"loss": 1.323,
|
| 370 |
+
"step": 24500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.455773598090025,
|
| 374 |
+
"grad_norm": 4.78125,
|
| 375 |
+
"learning_rate": 2.5737106698499585e-05,
|
| 376 |
+
"loss": 1.3206,
|
| 377 |
+
"step": 25000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.4848890700518256,
|
| 381 |
+
"grad_norm": 5.0625,
|
| 382 |
+
"learning_rate": 2.5251848832469577e-05,
|
| 383 |
+
"loss": 1.319,
|
| 384 |
+
"step": 25500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.5140045420136259,
|
| 388 |
+
"grad_norm": 5.34375,
|
| 389 |
+
"learning_rate": 2.476659096643957e-05,
|
| 390 |
+
"loss": 1.3129,
|
| 391 |
+
"step": 26000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.5431200139754266,
|
| 395 |
+
"grad_norm": 4.96875,
|
| 396 |
+
"learning_rate": 2.428133310040956e-05,
|
| 397 |
+
"loss": 1.3163,
|
| 398 |
+
"step": 26500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.572235485937227,
|
| 402 |
+
"grad_norm": 4.59375,
|
| 403 |
+
"learning_rate": 2.3796075234379552e-05,
|
| 404 |
+
"loss": 1.313,
|
| 405 |
+
"step": 27000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.6013509578990277,
|
| 409 |
+
"grad_norm": 4.40625,
|
| 410 |
+
"learning_rate": 2.331081736834954e-05,
|
| 411 |
+
"loss": 1.3167,
|
| 412 |
+
"step": 27500
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 1.630466429860828,
|
| 416 |
+
"grad_norm": 5.09375,
|
| 417 |
+
"learning_rate": 2.282555950231953e-05,
|
| 418 |
+
"loss": 1.3151,
|
| 419 |
+
"step": 28000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.6595819018226285,
|
| 423 |
+
"grad_norm": 5.09375,
|
| 424 |
+
"learning_rate": 2.2340301636289527e-05,
|
| 425 |
+
"loss": 1.312,
|
| 426 |
+
"step": 28500
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 1.688697373784429,
|
| 430 |
+
"grad_norm": 4.71875,
|
| 431 |
+
"learning_rate": 2.185504377025952e-05,
|
| 432 |
+
"loss": 1.3196,
|
| 433 |
+
"step": 29000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.7178128457462294,
|
| 437 |
+
"grad_norm": 4.46875,
|
| 438 |
+
"learning_rate": 2.136978590422951e-05,
|
| 439 |
+
"loss": 1.3149,
|
| 440 |
+
"step": 29500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.7469283177080301,
|
| 444 |
+
"grad_norm": 4.71875,
|
| 445 |
+
"learning_rate": 2.0884528038199498e-05,
|
| 446 |
+
"loss": 1.3142,
|
| 447 |
+
"step": 30000
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.7469283177080301,
|
| 451 |
+
"eval_accuracy": 0.7175331317769543,
|
| 452 |
+
"eval_loss": 1.2938237190246582,
|
| 453 |
+
"eval_runtime": 6.453,
|
| 454 |
+
"eval_samples_per_second": 90.036,
|
| 455 |
+
"eval_steps_per_second": 2.944,
|
| 456 |
+
"step": 30000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.7760437896698305,
|
| 460 |
+
"grad_norm": 5.03125,
|
| 461 |
+
"learning_rate": 2.039927017216949e-05,
|
| 462 |
+
"loss": 1.3115,
|
| 463 |
+
"step": 30500
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.8051592616316312,
|
| 467 |
+
"grad_norm": 4.90625,
|
| 468 |
+
"learning_rate": 1.9914012306139485e-05,
|
| 469 |
+
"loss": 1.3155,
|
| 470 |
+
"step": 31000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.8342747335934315,
|
| 474 |
+
"grad_norm": 5.0,
|
| 475 |
+
"learning_rate": 1.9428754440109477e-05,
|
| 476 |
+
"loss": 1.3179,
|
| 477 |
+
"step": 31500
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.863390205555232,
|
| 481 |
+
"grad_norm": 5.21875,
|
| 482 |
+
"learning_rate": 1.8943496574079468e-05,
|
| 483 |
+
"loss": 1.3091,
|
| 484 |
+
"step": 32000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 1.8925056775170326,
|
| 488 |
+
"grad_norm": 4.46875,
|
| 489 |
+
"learning_rate": 1.8458238708049457e-05,
|
| 490 |
+
"loss": 1.3148,
|
| 491 |
+
"step": 32500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.921621149478833,
|
| 495 |
+
"grad_norm": 5.15625,
|
| 496 |
+
"learning_rate": 1.7972980842019448e-05,
|
| 497 |
+
"loss": 1.3091,
|
| 498 |
+
"step": 33000
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.9507366214406336,
|
| 502 |
+
"grad_norm": 5.03125,
|
| 503 |
+
"learning_rate": 1.7487722975989443e-05,
|
| 504 |
+
"loss": 1.3176,
|
| 505 |
+
"step": 33500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.979852093402434,
|
| 509 |
+
"grad_norm": 4.625,
|
| 510 |
+
"learning_rate": 1.7002465109959435e-05,
|
| 511 |
+
"loss": 1.3135,
|
| 512 |
+
"step": 34000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.0089675653642347,
|
| 516 |
+
"grad_norm": 5.03125,
|
| 517 |
+
"learning_rate": 1.6517207243929427e-05,
|
| 518 |
+
"loss": 1.3134,
|
| 519 |
+
"step": 34500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 2.038083037326035,
|
| 523 |
+
"grad_norm": 4.875,
|
| 524 |
+
"learning_rate": 1.6031949377899415e-05,
|
| 525 |
+
"loss": 1.3061,
|
| 526 |
+
"step": 35000
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.0671985092878353,
|
| 530 |
+
"grad_norm": 4.84375,
|
| 531 |
+
"learning_rate": 1.5546691511869406e-05,
|
| 532 |
+
"loss": 1.3053,
|
| 533 |
+
"step": 35500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.096313981249636,
|
| 537 |
+
"grad_norm": 4.40625,
|
| 538 |
+
"learning_rate": 1.5061433645839398e-05,
|
| 539 |
+
"loss": 1.3102,
|
| 540 |
+
"step": 36000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.1254294532114364,
|
| 544 |
+
"grad_norm": 4.90625,
|
| 545 |
+
"learning_rate": 1.4576175779809393e-05,
|
| 546 |
+
"loss": 1.3137,
|
| 547 |
+
"step": 36500
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 2.154544925173237,
|
| 551 |
+
"grad_norm": 4.75,
|
| 552 |
+
"learning_rate": 1.4090917913779383e-05,
|
| 553 |
+
"loss": 1.306,
|
| 554 |
+
"step": 37000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 2.1836603971350375,
|
| 558 |
+
"grad_norm": 4.9375,
|
| 559 |
+
"learning_rate": 1.3605660047749375e-05,
|
| 560 |
+
"loss": 1.3029,
|
| 561 |
+
"step": 37500
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.212775869096838,
|
| 565 |
+
"grad_norm": 5.34375,
|
| 566 |
+
"learning_rate": 1.3120402181719366e-05,
|
| 567 |
+
"loss": 1.3068,
|
| 568 |
+
"step": 38000
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.2418913410586385,
|
| 572 |
+
"grad_norm": 4.96875,
|
| 573 |
+
"learning_rate": 1.2635144315689356e-05,
|
| 574 |
+
"loss": 1.3033,
|
| 575 |
+
"step": 38500
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.2710068130204393,
|
| 579 |
+
"grad_norm": 5.46875,
|
| 580 |
+
"learning_rate": 1.214988644965935e-05,
|
| 581 |
+
"loss": 1.3104,
|
| 582 |
+
"step": 39000
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.3001222849822396,
|
| 586 |
+
"grad_norm": 4.71875,
|
| 587 |
+
"learning_rate": 1.1664628583629341e-05,
|
| 588 |
+
"loss": 1.3058,
|
| 589 |
+
"step": 39500
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 2.32923775694404,
|
| 593 |
+
"grad_norm": 4.53125,
|
| 594 |
+
"learning_rate": 1.1179370717599333e-05,
|
| 595 |
+
"loss": 1.3064,
|
| 596 |
+
"step": 40000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 2.32923775694404,
|
| 600 |
+
"eval_accuracy": 0.7175919626253005,
|
| 601 |
+
"eval_loss": 1.2843679189682007,
|
| 602 |
+
"eval_runtime": 12.4622,
|
| 603 |
+
"eval_samples_per_second": 46.621,
|
| 604 |
+
"eval_steps_per_second": 1.525,
|
| 605 |
+
"step": 40000
|
| 606 |
+
}
|
| 607 |
+
],
|
| 608 |
+
"logging_steps": 500,
|
| 609 |
+
"max_steps": 51519,
|
| 610 |
+
"num_input_tokens_seen": 0,
|
| 611 |
+
"num_train_epochs": 3,
|
| 612 |
+
"save_steps": 10000,
|
| 613 |
+
"total_flos": 2.5163609407488e+18,
|
| 614 |
+
"train_batch_size": 16,
|
| 615 |
+
"trial_name": null,
|
| 616 |
+
"trial_params": null
|
| 617 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b72e646bb40995a7e3ddeabf4857d1aebbd0ce8672eb70c45afd1146be1ff620
|
| 3 |
+
size 5048
|