Upload 18 files
Browse files- config.json +32 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- rng_state_2.pth +3 -0
- rng_state_3.pth +3 -0
- rng_state_4.pth +3 -0
- rng_state_5.pth +3 -0
- rng_state_6.pth +3 -0
- rng_state_7.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +44 -0
- trainer_state.json +383 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"pad_token_id": 3,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_embd": 384,
|
| 15 |
+
"n_head": 12,
|
| 16 |
+
"n_inner": null,
|
| 17 |
+
"n_layer": 12,
|
| 18 |
+
"n_positions": 256,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.43.4",
|
| 30 |
+
"use_cache": true,
|
| 31 |
+
"vocab_size": 12000
|
| 32 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 3,
|
| 5 |
+
"transformers_version": "4.43.4"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e083606c4420f6b0eabf7f7f54dd4f756722032ac6d1e1226d5a6cb97d4cceaf
|
| 3 |
+
size 104017304
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:129f26b82d50287299e3cf4364237ac2e53bed00156369dd18150b40590bb282
|
| 3 |
+
size 208128954
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:81200223bf2c0d600698970a7c2ba66b5b0790b87726e5859fceea4f062c23f0
|
| 3 |
+
size 15984
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:43a94eca64adf682e6c4afdff5a761218f2ad867734f252ae336a3322875bf3a
|
| 3 |
+
size 15984
|
rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8baf5eb40487de162d10fc74bfdac50b5d4102978b796d3cbdaf051869514398
|
| 3 |
+
size 15984
|
rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75707fde14da5c30bdf811143e8c5947e66120b7bfb5905ca97ae767c0edea1a
|
| 3 |
+
size 15984
|
rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:194c2c331d205afd9dd93d5ae00960df17a31763d26e08bad72c6c0628c815da
|
| 3 |
+
size 15984
|
rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b0354cf0563ccef647bf17d17da90532c48be4003332acea51bf66ac168630c8
|
| 3 |
+
size 15984
|
rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6831753ddb3c34fda85ead6b6ffa64bcbce1aa2bb0154b33350d078b298b59f2
|
| 3 |
+
size 15984
|
rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47d2bcbd3fa196b84438a458da4848bbe2b3085a365e7e96466ddce11dcc6378
|
| 3 |
+
size 15984
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b4aa4413016252ded4c541d27453552fd8f4998404969413cac614394667153
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|bos|>",
|
| 3 |
+
"eos_token": "<|eos|>",
|
| 4 |
+
"mask_token": "<|mask|>",
|
| 5 |
+
"pad_token": "<|pad|>"
|
| 6 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<|mask|>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<|bos|>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "<|eos|>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<|pad|>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"bos_token": "<|bos|>",
|
| 37 |
+
"clean_up_tokenization_spaces": true,
|
| 38 |
+
"eos_token": "<|eos|>",
|
| 39 |
+
"mask_token": "<|mask|>",
|
| 40 |
+
"model_max_length": 256,
|
| 41 |
+
"pad_token": "<|pad|>",
|
| 42 |
+
"return_tensors": "pt",
|
| 43 |
+
"tokenizer_class": "PreTrainedTokenizerFast"
|
| 44 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,383 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 8.330556481172943,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 25000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.16661112962345884,
|
| 13 |
+
"grad_norm": 2.0231502056121826,
|
| 14 |
+
"learning_rate": 2.776852160390981e-06,
|
| 15 |
+
"loss": 8.4699,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.3332222592469177,
|
| 20 |
+
"grad_norm": 1.640729546546936,
|
| 21 |
+
"learning_rate": 5.553704320781962e-06,
|
| 22 |
+
"loss": 7.446,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.49983338887037654,
|
| 27 |
+
"grad_norm": 1.6770328283309937,
|
| 28 |
+
"learning_rate": 8.330556481172942e-06,
|
| 29 |
+
"loss": 6.6034,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.6664445184938353,
|
| 34 |
+
"grad_norm": 2.338146924972534,
|
| 35 |
+
"learning_rate": 1.1107408641563923e-05,
|
| 36 |
+
"loss": 5.9787,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.8330556481172943,
|
| 41 |
+
"grad_norm": 2.557213068008423,
|
| 42 |
+
"learning_rate": 1.3884260801954905e-05,
|
| 43 |
+
"loss": 5.5732,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.9996667777407531,
|
| 48 |
+
"grad_norm": 1.7546786069869995,
|
| 49 |
+
"learning_rate": 1.6661112962345883e-05,
|
| 50 |
+
"loss": 5.2881,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.166277907364212,
|
| 55 |
+
"grad_norm": 1.7593083381652832,
|
| 56 |
+
"learning_rate": 1.9437965122736867e-05,
|
| 57 |
+
"loss": 5.0771,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.3328890369876707,
|
| 62 |
+
"grad_norm": 1.860612392425537,
|
| 63 |
+
"learning_rate": 2.2214817283127847e-05,
|
| 64 |
+
"loss": 4.8991,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.4995001666111296,
|
| 69 |
+
"grad_norm": 2.036371946334839,
|
| 70 |
+
"learning_rate": 2.4991669443518826e-05,
|
| 71 |
+
"loss": 4.7465,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.6661112962345883,
|
| 76 |
+
"grad_norm": 1.8470864295959473,
|
| 77 |
+
"learning_rate": 2.776852160390981e-05,
|
| 78 |
+
"loss": 4.5925,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.8327224258580475,
|
| 83 |
+
"grad_norm": 1.9861804246902466,
|
| 84 |
+
"learning_rate": 3.0545373764300786e-05,
|
| 85 |
+
"loss": 4.4767,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.9993335554815062,
|
| 90 |
+
"grad_norm": 2.018094062805176,
|
| 91 |
+
"learning_rate": 3.3322225924691766e-05,
|
| 92 |
+
"loss": 4.3708,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 2.165944685104965,
|
| 97 |
+
"grad_norm": 1.8410370349884033,
|
| 98 |
+
"learning_rate": 3.609352438076197e-05,
|
| 99 |
+
"loss": 4.2741,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 2.332555814728424,
|
| 104 |
+
"grad_norm": 1.7712337970733643,
|
| 105 |
+
"learning_rate": 3.887037654115295e-05,
|
| 106 |
+
"loss": 4.192,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 2.4991669443518827,
|
| 111 |
+
"grad_norm": 1.7352423667907715,
|
| 112 |
+
"learning_rate": 4.164722870154393e-05,
|
| 113 |
+
"loss": 4.1192,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 2.6657780739753414,
|
| 118 |
+
"grad_norm": 1.8459163904190063,
|
| 119 |
+
"learning_rate": 4.442408086193491e-05,
|
| 120 |
+
"loss": 4.0351,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 2.8323892035988005,
|
| 125 |
+
"grad_norm": 2.009974956512451,
|
| 126 |
+
"learning_rate": 4.719537931800511e-05,
|
| 127 |
+
"loss": 3.9709,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.9990003332222592,
|
| 132 |
+
"grad_norm": 1.7068768739700317,
|
| 133 |
+
"learning_rate": 4.997223147839609e-05,
|
| 134 |
+
"loss": 3.899,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 3.165611462845718,
|
| 139 |
+
"grad_norm": 1.6234753131866455,
|
| 140 |
+
"learning_rate": 5.274908363878708e-05,
|
| 141 |
+
"loss": 3.8237,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 3.332222592469177,
|
| 146 |
+
"grad_norm": 1.786528468132019,
|
| 147 |
+
"learning_rate": 5.552593579917805e-05,
|
| 148 |
+
"loss": 3.7704,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 3.4988337220926358,
|
| 153 |
+
"grad_norm": 1.7539526224136353,
|
| 154 |
+
"learning_rate": 5.829723425524826e-05,
|
| 155 |
+
"loss": 3.7183,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 3.665444851716095,
|
| 160 |
+
"grad_norm": 1.6715501546859741,
|
| 161 |
+
"learning_rate": 6.107408641563923e-05,
|
| 162 |
+
"loss": 3.6671,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 3.8320559813395536,
|
| 167 |
+
"grad_norm": 1.7561357021331787,
|
| 168 |
+
"learning_rate": 6.385093857603022e-05,
|
| 169 |
+
"loss": 3.6214,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.9986671109630123,
|
| 174 |
+
"grad_norm": 1.5936201810836792,
|
| 175 |
+
"learning_rate": 6.662779073642119e-05,
|
| 176 |
+
"loss": 3.5782,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 4.1652782405864714,
|
| 181 |
+
"grad_norm": 1.7363481521606445,
|
| 182 |
+
"learning_rate": 6.940464289681218e-05,
|
| 183 |
+
"loss": 3.5291,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 4.33188937020993,
|
| 188 |
+
"grad_norm": 1.5552008152008057,
|
| 189 |
+
"learning_rate": 7.217594135288238e-05,
|
| 190 |
+
"loss": 3.5017,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 4.498500499833389,
|
| 195 |
+
"grad_norm": 1.4608962535858154,
|
| 196 |
+
"learning_rate": 7.495279351327336e-05,
|
| 197 |
+
"loss": 3.4685,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 4.665111629456848,
|
| 202 |
+
"grad_norm": 1.470942735671997,
|
| 203 |
+
"learning_rate": 7.772964567366434e-05,
|
| 204 |
+
"loss": 3.4407,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 4.831722759080306,
|
| 209 |
+
"grad_norm": 1.4744900465011597,
|
| 210 |
+
"learning_rate": 8.050649783405532e-05,
|
| 211 |
+
"loss": 3.4156,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 4.998333888703765,
|
| 216 |
+
"grad_norm": 1.449549913406372,
|
| 217 |
+
"learning_rate": 8.32833499944463e-05,
|
| 218 |
+
"loss": 3.3933,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 5.1649450183272245,
|
| 223 |
+
"grad_norm": 1.4103004932403564,
|
| 224 |
+
"learning_rate": 8.606020215483728e-05,
|
| 225 |
+
"loss": 3.355,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 5.331556147950683,
|
| 230 |
+
"grad_norm": 1.3490815162658691,
|
| 231 |
+
"learning_rate": 8.883705431522826e-05,
|
| 232 |
+
"loss": 3.335,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 5.498167277574142,
|
| 237 |
+
"grad_norm": 1.3762222528457642,
|
| 238 |
+
"learning_rate": 9.161390647561924e-05,
|
| 239 |
+
"loss": 3.3183,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 5.664778407197601,
|
| 244 |
+
"grad_norm": 1.3008846044540405,
|
| 245 |
+
"learning_rate": 9.438520493168944e-05,
|
| 246 |
+
"loss": 3.3013,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 5.831389536821059,
|
| 251 |
+
"grad_norm": 1.3132846355438232,
|
| 252 |
+
"learning_rate": 9.716205709208043e-05,
|
| 253 |
+
"loss": 3.2846,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 5.9980006664445185,
|
| 258 |
+
"grad_norm": 1.194407343864441,
|
| 259 |
+
"learning_rate": 9.99389092524714e-05,
|
| 260 |
+
"loss": 3.2706,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 6.164611796067978,
|
| 265 |
+
"grad_norm": 1.1745057106018066,
|
| 266 |
+
"learning_rate": 9.996657887827809e-05,
|
| 267 |
+
"loss": 3.2324,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 6.331222925691436,
|
| 272 |
+
"grad_norm": 1.1769245862960815,
|
| 273 |
+
"learning_rate": 9.986361454903414e-05,
|
| 274 |
+
"loss": 3.2237,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 6.497834055314895,
|
| 279 |
+
"grad_norm": 1.1529046297073364,
|
| 280 |
+
"learning_rate": 9.969082137309689e-05,
|
| 281 |
+
"loss": 3.2103,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 6.664445184938354,
|
| 286 |
+
"grad_norm": 1.1225526332855225,
|
| 287 |
+
"learning_rate": 9.944857739312099e-05,
|
| 288 |
+
"loss": 3.1936,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 6.831056314561812,
|
| 293 |
+
"grad_norm": 1.1167833805084229,
|
| 294 |
+
"learning_rate": 9.913722118348934e-05,
|
| 295 |
+
"loss": 3.1818,
|
| 296 |
+
"step": 20500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 6.9976674441852715,
|
| 300 |
+
"grad_norm": 1.093309998512268,
|
| 301 |
+
"learning_rate": 9.875801618300155e-05,
|
| 302 |
+
"loss": 3.1733,
|
| 303 |
+
"step": 21000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 7.164278573808731,
|
| 307 |
+
"grad_norm": 1.1539673805236816,
|
| 308 |
+
"learning_rate": 9.83099727084671e-05,
|
| 309 |
+
"loss": 3.1342,
|
| 310 |
+
"step": 21500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 7.330889703432189,
|
| 314 |
+
"grad_norm": 1.1930997371673584,
|
| 315 |
+
"learning_rate": 9.779440838491111e-05,
|
| 316 |
+
"loss": 3.1356,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 7.497500833055648,
|
| 321 |
+
"grad_norm": 1.0827438831329346,
|
| 322 |
+
"learning_rate": 9.721204379524642e-05,
|
| 323 |
+
"loss": 3.1231,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 7.664111962679107,
|
| 328 |
+
"grad_norm": 1.0850555896759033,
|
| 329 |
+
"learning_rate": 9.656505485544595e-05,
|
| 330 |
+
"loss": 3.1126,
|
| 331 |
+
"step": 23000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 7.8307230923025655,
|
| 335 |
+
"grad_norm": 1.0119590759277344,
|
| 336 |
+
"learning_rate": 9.585175299579849e-05,
|
| 337 |
+
"loss": 3.1049,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 7.997334221926025,
|
| 342 |
+
"grad_norm": 1.0027412176132202,
|
| 343 |
+
"learning_rate": 9.507436603906176e-05,
|
| 344 |
+
"loss": 3.1004,
|
| 345 |
+
"step": 24000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 8.163945351549483,
|
| 349 |
+
"grad_norm": 1.0899256467819214,
|
| 350 |
+
"learning_rate": 9.423398050680459e-05,
|
| 351 |
+
"loss": 3.0591,
|
| 352 |
+
"step": 24500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 8.330556481172943,
|
| 356 |
+
"grad_norm": 1.0424436330795288,
|
| 357 |
+
"learning_rate": 9.333363626660517e-05,
|
| 358 |
+
"loss": 3.0582,
|
| 359 |
+
"step": 25000
|
| 360 |
+
}
|
| 361 |
+
],
|
| 362 |
+
"logging_steps": 500,
|
| 363 |
+
"max_steps": 60020,
|
| 364 |
+
"num_input_tokens_seen": 0,
|
| 365 |
+
"num_train_epochs": 20,
|
| 366 |
+
"save_steps": 1000,
|
| 367 |
+
"stateful_callbacks": {
|
| 368 |
+
"TrainerControl": {
|
| 369 |
+
"args": {
|
| 370 |
+
"should_epoch_stop": false,
|
| 371 |
+
"should_evaluate": false,
|
| 372 |
+
"should_log": false,
|
| 373 |
+
"should_save": true,
|
| 374 |
+
"should_training_stop": false
|
| 375 |
+
},
|
| 376 |
+
"attributes": {}
|
| 377 |
+
}
|
| 378 |
+
},
|
| 379 |
+
"total_flos": 1.046659236626432e+17,
|
| 380 |
+
"train_batch_size": 16,
|
| 381 |
+
"trial_name": null,
|
| 382 |
+
"trial_params": null
|
| 383 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:90e9c2c1a12362f57e5abb4006a9e6b3f97922c1053722b8e4b5e741a1a2790d
|
| 3 |
+
size 5240
|