Upload folder using huggingface_hub
Browse files- all_results.json +9 -0
- config.json +30 -0
- generation_config.json +7 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +23 -0
- tokenizer.json +0 -0
- tokenizer_config.json +23 -0
- train_results.json +9 -0
- trainer_state.json +467 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 1.543347550373806e+17,
|
| 4 |
+
"train_loss": 4.46121194826077,
|
| 5 |
+
"train_runtime": 2449.295,
|
| 6 |
+
"train_samples": 109262,
|
| 7 |
+
"train_samples_per_second": 178.438,
|
| 8 |
+
"train_steps_per_second": 1.395
|
| 9 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"head_dim": 96,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"max_position_embeddings": 1024,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 8,
|
| 18 |
+
"num_hidden_layers": 2,
|
| 19 |
+
"num_key_value_heads": 8,
|
| 20 |
+
"pad_token_id": 0,
|
| 21 |
+
"pretraining_tp": 1,
|
| 22 |
+
"rms_norm_eps": 1e-06,
|
| 23 |
+
"rope_scaling": null,
|
| 24 |
+
"rope_theta": 10000.0,
|
| 25 |
+
"tie_word_embeddings": false,
|
| 26 |
+
"torch_dtype": "float32",
|
| 27 |
+
"transformers_version": "4.48.0",
|
| 28 |
+
"use_cache": true,
|
| 29 |
+
"vocab_size": 50257
|
| 30 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.48.0"
|
| 7 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b80536138fea3ec6c5774693ce33f4d9178b04c64c2391321b5bd61ca3e9551
|
| 3 |
+
size 384294152
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"unk_token": {
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
}
|
| 23 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"50256": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
}
|
| 13 |
+
},
|
| 14 |
+
"bos_token": "<|endoftext|>",
|
| 15 |
+
"clean_up_tokenization_spaces": false,
|
| 16 |
+
"eos_token": "<|endoftext|>",
|
| 17 |
+
"errors": "replace",
|
| 18 |
+
"extra_special_tokens": {},
|
| 19 |
+
"model_max_length": 1024,
|
| 20 |
+
"pad_token": null,
|
| 21 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 22 |
+
"unk_token": "<|endoftext|>"
|
| 23 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 1.543347550373806e+17,
|
| 4 |
+
"train_loss": 4.46121194826077,
|
| 5 |
+
"train_runtime": 2449.295,
|
| 6 |
+
"train_samples": 109262,
|
| 7 |
+
"train_samples_per_second": 178.438,
|
| 8 |
+
"train_steps_per_second": 1.395
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,467 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 3416,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.117096018735363,
|
| 13 |
+
"eval_accuracy": 0.16272885212172436,
|
| 14 |
+
"eval_loss": 6.373325347900391,
|
| 15 |
+
"eval_runtime": 1.492,
|
| 16 |
+
"eval_samples_per_second": 153.483,
|
| 17 |
+
"eval_steps_per_second": 1.34,
|
| 18 |
+
"step": 100
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.234192037470726,
|
| 22 |
+
"grad_norm": 0.5273725986480713,
|
| 23 |
+
"learning_rate": 0.00029824050644124145,
|
| 24 |
+
"loss": 6.8543,
|
| 25 |
+
"step": 200
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.234192037470726,
|
| 29 |
+
"eval_accuracy": 0.19693341358364602,
|
| 30 |
+
"eval_loss": 5.696830749511719,
|
| 31 |
+
"eval_runtime": 1.1659,
|
| 32 |
+
"eval_samples_per_second": 196.414,
|
| 33 |
+
"eval_steps_per_second": 1.715,
|
| 34 |
+
"step": 200
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.351288056206089,
|
| 38 |
+
"eval_accuracy": 0.21933947162852643,
|
| 39 |
+
"eval_loss": 5.292416095733643,
|
| 40 |
+
"eval_runtime": 1.1779,
|
| 41 |
+
"eval_samples_per_second": 194.406,
|
| 42 |
+
"eval_steps_per_second": 1.698,
|
| 43 |
+
"step": 300
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.468384074941452,
|
| 47 |
+
"grad_norm": 0.7039262056350708,
|
| 48 |
+
"learning_rate": 0.00029145544445670097,
|
| 49 |
+
"loss": 5.3976,
|
| 50 |
+
"step": 400
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.468384074941452,
|
| 54 |
+
"eval_accuracy": 0.2335839021287676,
|
| 55 |
+
"eval_loss": 5.054598808288574,
|
| 56 |
+
"eval_runtime": 1.1808,
|
| 57 |
+
"eval_samples_per_second": 193.929,
|
| 58 |
+
"eval_steps_per_second": 1.694,
|
| 59 |
+
"step": 400
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.585480093676815,
|
| 63 |
+
"eval_accuracy": 0.24265474864150735,
|
| 64 |
+
"eval_loss": 4.891258716583252,
|
| 65 |
+
"eval_runtime": 1.1832,
|
| 66 |
+
"eval_samples_per_second": 193.538,
|
| 67 |
+
"eval_steps_per_second": 1.69,
|
| 68 |
+
"step": 500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.702576112412178,
|
| 72 |
+
"grad_norm": 0.7881871461868286,
|
| 73 |
+
"learning_rate": 0.0002797991505064349,
|
| 74 |
+
"loss": 4.9739,
|
| 75 |
+
"step": 600
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.702576112412178,
|
| 79 |
+
"eval_accuracy": 0.2506498994736775,
|
| 80 |
+
"eval_loss": 4.759438514709473,
|
| 81 |
+
"eval_runtime": 1.1726,
|
| 82 |
+
"eval_samples_per_second": 195.299,
|
| 83 |
+
"eval_steps_per_second": 1.706,
|
| 84 |
+
"step": 600
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.819672131147541,
|
| 88 |
+
"eval_accuracy": 0.2589993468990511,
|
| 89 |
+
"eval_loss": 4.651967525482178,
|
| 90 |
+
"eval_runtime": 1.1752,
|
| 91 |
+
"eval_samples_per_second": 194.864,
|
| 92 |
+
"eval_steps_per_second": 1.702,
|
| 93 |
+
"step": 700
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.936768149882904,
|
| 97 |
+
"grad_norm": 0.5959051251411438,
|
| 98 |
+
"learning_rate": 0.00026367302668500736,
|
| 99 |
+
"loss": 4.7436,
|
| 100 |
+
"step": 800
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.936768149882904,
|
| 104 |
+
"eval_accuracy": 0.2653638796757546,
|
| 105 |
+
"eval_loss": 4.566330432891846,
|
| 106 |
+
"eval_runtime": 1.1821,
|
| 107 |
+
"eval_samples_per_second": 193.727,
|
| 108 |
+
"eval_steps_per_second": 1.692,
|
| 109 |
+
"step": 800
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.053864168618267,
|
| 113 |
+
"eval_accuracy": 0.270379524218093,
|
| 114 |
+
"eval_loss": 4.497960567474365,
|
| 115 |
+
"eval_runtime": 1.1842,
|
| 116 |
+
"eval_samples_per_second": 193.385,
|
| 117 |
+
"eval_steps_per_second": 1.689,
|
| 118 |
+
"step": 900
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.17096018735363,
|
| 122 |
+
"grad_norm": 0.6118926405906677,
|
| 123 |
+
"learning_rate": 0.00024363240043240403,
|
| 124 |
+
"loss": 4.5595,
|
| 125 |
+
"step": 1000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.17096018735363,
|
| 129 |
+
"eval_accuracy": 0.2755872572748189,
|
| 130 |
+
"eval_loss": 4.44118070602417,
|
| 131 |
+
"eval_runtime": 1.1779,
|
| 132 |
+
"eval_samples_per_second": 194.409,
|
| 133 |
+
"eval_steps_per_second": 1.698,
|
| 134 |
+
"step": 1000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.288056206088993,
|
| 138 |
+
"eval_accuracy": 0.2790832682366701,
|
| 139 |
+
"eval_loss": 4.388593673706055,
|
| 140 |
+
"eval_runtime": 1.1719,
|
| 141 |
+
"eval_samples_per_second": 195.408,
|
| 142 |
+
"eval_steps_per_second": 1.707,
|
| 143 |
+
"step": 1100
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.405152224824356,
|
| 147 |
+
"grad_norm": 0.6060830354690552,
|
| 148 |
+
"learning_rate": 0.00022036740099454973,
|
| 149 |
+
"loss": 4.44,
|
| 150 |
+
"step": 1200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.405152224824356,
|
| 154 |
+
"eval_accuracy": 0.28338605095894853,
|
| 155 |
+
"eval_loss": 4.339376449584961,
|
| 156 |
+
"eval_runtime": 1.1734,
|
| 157 |
+
"eval_samples_per_second": 195.157,
|
| 158 |
+
"eval_steps_per_second": 1.704,
|
| 159 |
+
"step": 1200
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.5222482435597189,
|
| 163 |
+
"eval_accuracy": 0.28774005728506363,
|
| 164 |
+
"eval_loss": 4.292956352233887,
|
| 165 |
+
"eval_runtime": 1.1856,
|
| 166 |
+
"eval_samples_per_second": 193.144,
|
| 167 |
+
"eval_steps_per_second": 1.687,
|
| 168 |
+
"step": 1300
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.639344262295082,
|
| 172 |
+
"grad_norm": 0.5661224126815796,
|
| 173 |
+
"learning_rate": 0.0001946791937799555,
|
| 174 |
+
"loss": 4.3482,
|
| 175 |
+
"step": 1400
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.639344262295082,
|
| 179 |
+
"eval_accuracy": 0.2911549642075068,
|
| 180 |
+
"eval_loss": 4.25685977935791,
|
| 181 |
+
"eval_runtime": 1.175,
|
| 182 |
+
"eval_samples_per_second": 194.899,
|
| 183 |
+
"eval_steps_per_second": 1.702,
|
| 184 |
+
"step": 1400
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.756440281030445,
|
| 188 |
+
"eval_accuracy": 0.29346429501380905,
|
| 189 |
+
"eval_loss": 4.224252700805664,
|
| 190 |
+
"eval_runtime": 1.1755,
|
| 191 |
+
"eval_samples_per_second": 194.807,
|
| 192 |
+
"eval_steps_per_second": 1.701,
|
| 193 |
+
"step": 1500
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.8735362997658078,
|
| 197 |
+
"grad_norm": 0.5346641540527344,
|
| 198 |
+
"learning_rate": 0.00016745239101833442,
|
| 199 |
+
"loss": 4.2765,
|
| 200 |
+
"step": 1600
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 1.8735362997658078,
|
| 204 |
+
"eval_accuracy": 0.29705421591602743,
|
| 205 |
+
"eval_loss": 4.186337947845459,
|
| 206 |
+
"eval_runtime": 1.1767,
|
| 207 |
+
"eval_samples_per_second": 194.606,
|
| 208 |
+
"eval_steps_per_second": 1.7,
|
| 209 |
+
"step": 1600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.990632318501171,
|
| 213 |
+
"eval_accuracy": 0.3002471538885118,
|
| 214 |
+
"eval_loss": 4.158440113067627,
|
| 215 |
+
"eval_runtime": 1.1742,
|
| 216 |
+
"eval_samples_per_second": 195.024,
|
| 217 |
+
"eval_steps_per_second": 1.703,
|
| 218 |
+
"step": 1700
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.107728337236534,
|
| 222 |
+
"grad_norm": 0.5907180309295654,
|
| 223 |
+
"learning_rate": 0.0001396245888018435,
|
| 224 |
+
"loss": 4.1869,
|
| 225 |
+
"step": 1800
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.107728337236534,
|
| 229 |
+
"eval_accuracy": 0.3025009924573243,
|
| 230 |
+
"eval_loss": 4.138100624084473,
|
| 231 |
+
"eval_runtime": 1.1874,
|
| 232 |
+
"eval_samples_per_second": 192.866,
|
| 233 |
+
"eval_steps_per_second": 1.684,
|
| 234 |
+
"step": 1800
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.2248243559718968,
|
| 238 |
+
"eval_accuracy": 0.30516461985683,
|
| 239 |
+
"eval_loss": 4.114769458770752,
|
| 240 |
+
"eval_runtime": 1.1787,
|
| 241 |
+
"eval_samples_per_second": 194.278,
|
| 242 |
+
"eval_steps_per_second": 1.697,
|
| 243 |
+
"step": 1900
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.34192037470726,
|
| 247 |
+
"grad_norm": 0.5956910848617554,
|
| 248 |
+
"learning_rate": 0.00011215407954696412,
|
| 249 |
+
"loss": 4.1154,
|
| 250 |
+
"step": 2000
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.34192037470726,
|
| 254 |
+
"eval_accuracy": 0.30651777672484815,
|
| 255 |
+
"eval_loss": 4.09398889541626,
|
| 256 |
+
"eval_runtime": 1.1823,
|
| 257 |
+
"eval_samples_per_second": 193.693,
|
| 258 |
+
"eval_steps_per_second": 1.692,
|
| 259 |
+
"step": 2000
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 2.459016393442623,
|
| 263 |
+
"eval_accuracy": 0.30912164325321106,
|
| 264 |
+
"eval_loss": 4.074470043182373,
|
| 265 |
+
"eval_runtime": 1.1705,
|
| 266 |
+
"eval_samples_per_second": 195.64,
|
| 267 |
+
"eval_steps_per_second": 1.709,
|
| 268 |
+
"step": 2100
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 2.576112412177986,
|
| 272 |
+
"grad_norm": 0.6125728487968445,
|
| 273 |
+
"learning_rate": 8.611292881304631e-05,
|
| 274 |
+
"loss": 4.0822,
|
| 275 |
+
"step": 2200
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 2.576112412177986,
|
| 279 |
+
"eval_accuracy": 0.31070530633849414,
|
| 280 |
+
"eval_loss": 4.0601091384887695,
|
| 281 |
+
"eval_runtime": 1.176,
|
| 282 |
+
"eval_samples_per_second": 194.735,
|
| 283 |
+
"eval_steps_per_second": 1.701,
|
| 284 |
+
"step": 2200
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.693208430913349,
|
| 288 |
+
"eval_accuracy": 0.31145231722777855,
|
| 289 |
+
"eval_loss": 4.047569274902344,
|
| 290 |
+
"eval_runtime": 1.1721,
|
| 291 |
+
"eval_samples_per_second": 195.374,
|
| 292 |
+
"eval_steps_per_second": 1.706,
|
| 293 |
+
"step": 2300
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.810304449648712,
|
| 297 |
+
"grad_norm": 0.5131784081459045,
|
| 298 |
+
"learning_rate": 6.21369402825085e-05,
|
| 299 |
+
"loss": 4.0496,
|
| 300 |
+
"step": 2400
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 2.810304449648712,
|
| 304 |
+
"eval_accuracy": 0.3138342148061827,
|
| 305 |
+
"eval_loss": 4.032547950744629,
|
| 306 |
+
"eval_runtime": 1.1719,
|
| 307 |
+
"eval_samples_per_second": 195.415,
|
| 308 |
+
"eval_steps_per_second": 1.707,
|
| 309 |
+
"step": 2400
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.927400468384075,
|
| 313 |
+
"eval_accuracy": 0.31531969931744547,
|
| 314 |
+
"eval_loss": 4.0211310386657715,
|
| 315 |
+
"eval_runtime": 1.1766,
|
| 316 |
+
"eval_samples_per_second": 194.63,
|
| 317 |
+
"eval_steps_per_second": 1.7,
|
| 318 |
+
"step": 2500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.0444964871194378,
|
| 322 |
+
"grad_norm": 0.5091509819030762,
|
| 323 |
+
"learning_rate": 4.118664897236514e-05,
|
| 324 |
+
"loss": 4.0193,
|
| 325 |
+
"step": 2600
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.0444964871194378,
|
| 329 |
+
"eval_accuracy": 0.31634844002783147,
|
| 330 |
+
"eval_loss": 4.012126445770264,
|
| 331 |
+
"eval_runtime": 1.1748,
|
| 332 |
+
"eval_samples_per_second": 194.929,
|
| 333 |
+
"eval_steps_per_second": 1.702,
|
| 334 |
+
"step": 2600
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 3.161592505854801,
|
| 338 |
+
"eval_accuracy": 0.31667712481911664,
|
| 339 |
+
"eval_loss": 4.007138729095459,
|
| 340 |
+
"eval_runtime": 1.1742,
|
| 341 |
+
"eval_samples_per_second": 195.034,
|
| 342 |
+
"eval_steps_per_second": 1.703,
|
| 343 |
+
"step": 2700
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 3.278688524590164,
|
| 347 |
+
"grad_norm": 0.4789488911628723,
|
| 348 |
+
"learning_rate": 2.39835098192799e-05,
|
| 349 |
+
"loss": 3.9665,
|
| 350 |
+
"step": 2800
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 3.278688524590164,
|
| 354 |
+
"eval_accuracy": 0.31719362949113616,
|
| 355 |
+
"eval_loss": 4.001045227050781,
|
| 356 |
+
"eval_runtime": 1.1742,
|
| 357 |
+
"eval_samples_per_second": 195.029,
|
| 358 |
+
"eval_steps_per_second": 1.703,
|
| 359 |
+
"step": 2800
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.3957845433255267,
|
| 363 |
+
"eval_accuracy": 0.31759488105452327,
|
| 364 |
+
"eval_loss": 3.997783899307251,
|
| 365 |
+
"eval_runtime": 1.1723,
|
| 366 |
+
"eval_samples_per_second": 195.344,
|
| 367 |
+
"eval_steps_per_second": 1.706,
|
| 368 |
+
"step": 2900
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.51288056206089,
|
| 372 |
+
"grad_norm": 0.4739971458911896,
|
| 373 |
+
"learning_rate": 1.1119938913529075e-05,
|
| 374 |
+
"loss": 3.9575,
|
| 375 |
+
"step": 3000
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.51288056206089,
|
| 379 |
+
"eval_accuracy": 0.3181796838649917,
|
| 380 |
+
"eval_loss": 3.993601083755493,
|
| 381 |
+
"eval_runtime": 1.1865,
|
| 382 |
+
"eval_samples_per_second": 192.998,
|
| 383 |
+
"eval_steps_per_second": 1.686,
|
| 384 |
+
"step": 3000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 3.629976580796253,
|
| 388 |
+
"eval_accuracy": 0.3181924897659508,
|
| 389 |
+
"eval_loss": 3.991323947906494,
|
| 390 |
+
"eval_runtime": 1.1742,
|
| 391 |
+
"eval_samples_per_second": 195.026,
|
| 392 |
+
"eval_steps_per_second": 1.703,
|
| 393 |
+
"step": 3100
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 3.747072599531616,
|
| 397 |
+
"grad_norm": 0.47446030378341675,
|
| 398 |
+
"learning_rate": 3.038912755806222e-06,
|
| 399 |
+
"loss": 3.9558,
|
| 400 |
+
"step": 3200
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 3.747072599531616,
|
| 404 |
+
"eval_accuracy": 0.3182010270332569,
|
| 405 |
+
"eval_loss": 3.9895448684692383,
|
| 406 |
+
"eval_runtime": 1.1801,
|
| 407 |
+
"eval_samples_per_second": 194.046,
|
| 408 |
+
"eval_steps_per_second": 1.695,
|
| 409 |
+
"step": 3200
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 3.8641686182669788,
|
| 413 |
+
"eval_accuracy": 0.3184998313889707,
|
| 414 |
+
"eval_loss": 3.9884603023529053,
|
| 415 |
+
"eval_runtime": 1.1735,
|
| 416 |
+
"eval_samples_per_second": 195.146,
|
| 417 |
+
"eval_steps_per_second": 1.704,
|
| 418 |
+
"step": 3300
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 3.981264637002342,
|
| 422 |
+
"grad_norm": 0.4558132290840149,
|
| 423 |
+
"learning_rate": 1.8713692093474952e-08,
|
| 424 |
+
"loss": 3.9531,
|
| 425 |
+
"step": 3400
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 3.981264637002342,
|
| 429 |
+
"eval_accuracy": 0.3183333546765016,
|
| 430 |
+
"eval_loss": 3.988285541534424,
|
| 431 |
+
"eval_runtime": 1.1729,
|
| 432 |
+
"eval_samples_per_second": 195.237,
|
| 433 |
+
"eval_steps_per_second": 1.705,
|
| 434 |
+
"step": 3400
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.0,
|
| 438 |
+
"step": 3416,
|
| 439 |
+
"total_flos": 1.543347550373806e+17,
|
| 440 |
+
"train_loss": 4.46121194826077,
|
| 441 |
+
"train_runtime": 2449.295,
|
| 442 |
+
"train_samples_per_second": 178.438,
|
| 443 |
+
"train_steps_per_second": 1.395
|
| 444 |
+
}
|
| 445 |
+
],
|
| 446 |
+
"logging_steps": 200,
|
| 447 |
+
"max_steps": 3416,
|
| 448 |
+
"num_input_tokens_seen": 0,
|
| 449 |
+
"num_train_epochs": 4,
|
| 450 |
+
"save_steps": 5000,
|
| 451 |
+
"stateful_callbacks": {
|
| 452 |
+
"TrainerControl": {
|
| 453 |
+
"args": {
|
| 454 |
+
"should_epoch_stop": false,
|
| 455 |
+
"should_evaluate": false,
|
| 456 |
+
"should_log": false,
|
| 457 |
+
"should_save": true,
|
| 458 |
+
"should_training_stop": true
|
| 459 |
+
},
|
| 460 |
+
"attributes": {}
|
| 461 |
+
}
|
| 462 |
+
},
|
| 463 |
+
"total_flos": 1.543347550373806e+17,
|
| 464 |
+
"train_batch_size": 128,
|
| 465 |
+
"trial_name": null,
|
| 466 |
+
"trial_params": null
|
| 467 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1c5d3c134fe5baca62831c31d5ed9ef7b535662cae7f7a1e65982d7088b1888
|
| 3 |
+
size 5304
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|