Upload folder using huggingface_hub
Browse files- config.json +34 -0
- generation_config.json +12 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +20 -0
- trainer_state.json +790 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"add_cross_attention": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"dtype": "float32",
|
| 10 |
+
"embd_pdrop": 0.1,
|
| 11 |
+
"eos_token_id": 50256,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"layer_norm_epsilon": 1e-05,
|
| 14 |
+
"model_type": "gpt2",
|
| 15 |
+
"n_embd": 384,
|
| 16 |
+
"n_head": 6,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 6,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"pad_token_id": 50256,
|
| 21 |
+
"reorder_and_upcast_attn": false,
|
| 22 |
+
"resid_pdrop": 0.1,
|
| 23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 24 |
+
"scale_attn_weights": true,
|
| 25 |
+
"summary_activation": null,
|
| 26 |
+
"summary_first_dropout": 0.1,
|
| 27 |
+
"summary_proj_to_labels": true,
|
| 28 |
+
"summary_type": "cls_index",
|
| 29 |
+
"summary_use_proj": true,
|
| 30 |
+
"tie_word_embeddings": true,
|
| 31 |
+
"transformers_version": "5.10.2",
|
| 32 |
+
"use_cache": false,
|
| 33 |
+
"vocab_size": 50257
|
| 34 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
50256
|
| 6 |
+
],
|
| 7 |
+
"output_attentions": false,
|
| 8 |
+
"output_hidden_states": false,
|
| 9 |
+
"pad_token_id": 50256,
|
| 10 |
+
"transformers_version": "5.10.2",
|
| 11 |
+
"use_cache": false
|
| 12 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:445a8f535479c6e2390770667d510481570e62662dc5df0fa5aa488a24d610d5
|
| 3 |
+
size 121365368
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a215bd10cebf6046528baa308b8e62f3c705fdd86732521894690352a8f8ea2
|
| 3 |
+
size 242781003
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b43faa323fedb496ca92fc659d81d3f6dc2f73f65b0899c8580623322a05027
|
| 3 |
+
size 14645
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0563f0c122c10b3bb9fb8ef7a029cf4fe15b88c3307945ff50155d3485ee7edd
|
| 3 |
+
size 1383
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c46a133444f68fa3f85a06bdf6d8f77bc6b18d9650c14e76f5316a1c5eb9d6d9
|
| 3 |
+
size 1465
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"max_length": 512,
|
| 10 |
+
"model_max_length": 1024,
|
| 11 |
+
"pad_to_multiple_of": null,
|
| 12 |
+
"pad_token": "<|endoftext|>",
|
| 13 |
+
"pad_token_type_id": 0,
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"stride": 0,
|
| 16 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 17 |
+
"truncation_side": "right",
|
| 18 |
+
"truncation_strategy": "longest_first",
|
| 19 |
+
"unk_token": "<|endoftext|>"
|
| 20 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,790 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 4200,
|
| 3 |
+
"best_metric": 0.9021432995796204,
|
| 4 |
+
"best_model_checkpoint": "./chessslm-2000plus/checkpoint-4200",
|
| 5 |
+
"epoch": 0.6086074481959136,
|
| 6 |
+
"eval_steps": 200,
|
| 7 |
+
"global_step": 4200,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.007245326764237067,
|
| 14 |
+
"grad_norm": 0.28264278173446655,
|
| 15 |
+
"learning_rate": 2.45e-05,
|
| 16 |
+
"loss": 1.2346894073486328,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.014490653528474133,
|
| 21 |
+
"grad_norm": 0.25636932253837585,
|
| 22 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 23 |
+
"loss": 1.1820894622802733,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0217359802927112,
|
| 28 |
+
"grad_norm": 0.2744424343109131,
|
| 29 |
+
"learning_rate": 4.9993596199009453e-05,
|
| 30 |
+
"loss": 1.1531302642822265,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.028981307056948267,
|
| 35 |
+
"grad_norm": 0.28049495816230774,
|
| 36 |
+
"learning_rate": 4.997386280918375e-05,
|
| 37 |
+
"loss": 1.1511387634277344,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.028981307056948267,
|
| 42 |
+
"eval_loss": 1.1023718118667603,
|
| 43 |
+
"eval_runtime": 193.7165,
|
| 44 |
+
"eval_samples_per_second": 120.0,
|
| 45 |
+
"eval_steps_per_second": 3.753,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.036226633821185336,
|
| 50 |
+
"grad_norm": 0.2879795730113983,
|
| 51 |
+
"learning_rate": 4.994080768959056e-05,
|
| 52 |
+
"loss": 1.1302296447753906,
|
| 53 |
+
"step": 250
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.0434719605854224,
|
| 57 |
+
"grad_norm": 0.29017066955566406,
|
| 58 |
+
"learning_rate": 4.989444847271928e-05,
|
| 59 |
+
"loss": 1.1334593963623047,
|
| 60 |
+
"step": 300
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.05071728734965947,
|
| 64 |
+
"grad_norm": 0.31129732728004456,
|
| 65 |
+
"learning_rate": 4.98348098878224e-05,
|
| 66 |
+
"loss": 1.1211315155029298,
|
| 67 |
+
"step": 350
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.057962614113896534,
|
| 71 |
+
"grad_norm": 0.29630768299102783,
|
| 72 |
+
"learning_rate": 4.9761923747724296e-05,
|
| 73 |
+
"loss": 1.1175349426269532,
|
| 74 |
+
"step": 400
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.057962614113896534,
|
| 78 |
+
"eval_loss": 1.0756858587265015,
|
| 79 |
+
"eval_runtime": 193.6918,
|
| 80 |
+
"eval_samples_per_second": 120.015,
|
| 81 |
+
"eval_steps_per_second": 3.753,
|
| 82 |
+
"step": 400
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.0652079408781336,
|
| 86 |
+
"grad_norm": 0.2720015347003937,
|
| 87 |
+
"learning_rate": 4.9675828931851424e-05,
|
| 88 |
+
"loss": 1.1160638427734375,
|
| 89 |
+
"step": 450
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.07245326764237067,
|
| 93 |
+
"grad_norm": 0.29248911142349243,
|
| 94 |
+
"learning_rate": 4.9576571365492965e-05,
|
| 95 |
+
"loss": 1.1049542236328125,
|
| 96 |
+
"step": 500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.07969859440660773,
|
| 100 |
+
"grad_norm": 0.31138038635253906,
|
| 101 |
+
"learning_rate": 4.946420399530304e-05,
|
| 102 |
+
"loss": 1.1029164123535156,
|
| 103 |
+
"step": 550
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.0869439211708448,
|
| 107 |
+
"grad_norm": 0.2833043932914734,
|
| 108 |
+
"learning_rate": 4.933878676105756e-05,
|
| 109 |
+
"loss": 1.0982711791992188,
|
| 110 |
+
"step": 600
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.0869439211708448,
|
| 114 |
+
"eval_loss": 1.056943416595459,
|
| 115 |
+
"eval_runtime": 193.7026,
|
| 116 |
+
"eval_samples_per_second": 120.009,
|
| 117 |
+
"eval_steps_per_second": 3.753,
|
| 118 |
+
"step": 600
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.09418924793508188,
|
| 122 |
+
"grad_norm": 0.32613420486450195,
|
| 123 |
+
"learning_rate": 4.9200386563680734e-05,
|
| 124 |
+
"loss": 1.0998190307617188,
|
| 125 |
+
"step": 650
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.10143457469931894,
|
| 129 |
+
"grad_norm": 0.30005744099617004,
|
| 130 |
+
"learning_rate": 4.904907722955829e-05,
|
| 131 |
+
"loss": 1.0960050201416016,
|
| 132 |
+
"step": 700
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.10867990146355601,
|
| 136 |
+
"grad_norm": 0.3087840974330902,
|
| 137 |
+
"learning_rate": 4.8884939471156544e-05,
|
| 138 |
+
"loss": 1.0800698852539063,
|
| 139 |
+
"step": 750
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.11592522822779307,
|
| 143 |
+
"grad_norm": 0.3069000542163849,
|
| 144 |
+
"learning_rate": 4.8708060843968165e-05,
|
| 145 |
+
"loss": 1.088259506225586,
|
| 146 |
+
"step": 800
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.11592522822779307,
|
| 150 |
+
"eval_loss": 1.0397652387619019,
|
| 151 |
+
"eval_runtime": 193.5663,
|
| 152 |
+
"eval_samples_per_second": 120.093,
|
| 153 |
+
"eval_steps_per_second": 3.756,
|
| 154 |
+
"step": 800
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.12317055499203014,
|
| 158 |
+
"grad_norm": 0.3139374554157257,
|
| 159 |
+
"learning_rate": 4.8518535699807713e-05,
|
| 160 |
+
"loss": 1.0745314788818359,
|
| 161 |
+
"step": 850
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.1304158817562672,
|
| 165 |
+
"grad_norm": 0.29361429810523987,
|
| 166 |
+
"learning_rate": 4.831646513648189e-05,
|
| 167 |
+
"loss": 1.0730350494384766,
|
| 168 |
+
"step": 900
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.13766120852050429,
|
| 172 |
+
"grad_norm": 0.32080790400505066,
|
| 173 |
+
"learning_rate": 4.810195694386123e-05,
|
| 174 |
+
"loss": 1.0730361938476562,
|
| 175 |
+
"step": 950
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.14490653528474134,
|
| 179 |
+
"grad_norm": 0.3122475743293762,
|
| 180 |
+
"learning_rate": 4.787512554638205e-05,
|
| 181 |
+
"loss": 1.0709983825683593,
|
| 182 |
+
"step": 1000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.14490653528474134,
|
| 186 |
+
"eval_loss": 1.023197054862976,
|
| 187 |
+
"eval_runtime": 193.618,
|
| 188 |
+
"eval_samples_per_second": 120.061,
|
| 189 |
+
"eval_steps_per_second": 3.755,
|
| 190 |
+
"step": 1000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.1521518620489784,
|
| 194 |
+
"grad_norm": 0.33824747800827026,
|
| 195 |
+
"learning_rate": 4.763609194200942e-05,
|
| 196 |
+
"loss": 1.0684901428222657,
|
| 197 |
+
"step": 1050
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.15939718881321546,
|
| 201 |
+
"grad_norm": 0.30118507146835327,
|
| 202 |
+
"learning_rate": 4.7384983637693614e-05,
|
| 203 |
+
"loss": 1.0644412231445313,
|
| 204 |
+
"step": 1100
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.16664251557745255,
|
| 208 |
+
"grad_norm": 0.29680049419403076,
|
| 209 |
+
"learning_rate": 4.71219345813544e-05,
|
| 210 |
+
"loss": 1.0592123413085937,
|
| 211 |
+
"step": 1150
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.1738878423416896,
|
| 215 |
+
"grad_norm": 0.320403128862381,
|
| 216 |
+
"learning_rate": 4.684708509042971e-05,
|
| 217 |
+
"loss": 1.0508477020263671,
|
| 218 |
+
"step": 1200
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.1738878423416896,
|
| 222 |
+
"eval_loss": 1.0096791982650757,
|
| 223 |
+
"eval_runtime": 193.6179,
|
| 224 |
+
"eval_samples_per_second": 120.061,
|
| 225 |
+
"eval_steps_per_second": 3.755,
|
| 226 |
+
"step": 1200
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.18113316910592667,
|
| 230 |
+
"grad_norm": 0.3681316673755646,
|
| 231 |
+
"learning_rate": 4.656058177702647e-05,
|
| 232 |
+
"loss": 1.0457440185546876,
|
| 233 |
+
"step": 1250
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.18837849587016375,
|
| 237 |
+
"grad_norm": 0.3355785608291626,
|
| 238 |
+
"learning_rate": 4.626257746971382e-05,
|
| 239 |
+
"loss": 1.0470626068115234,
|
| 240 |
+
"step": 1300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1956238226344008,
|
| 244 |
+
"grad_norm": 0.339575856924057,
|
| 245 |
+
"learning_rate": 4.5953231132000175e-05,
|
| 246 |
+
"loss": 1.0423269653320313,
|
| 247 |
+
"step": 1350
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.20286914939863787,
|
| 251 |
+
"grad_norm": 0.33345919847488403,
|
| 252 |
+
"learning_rate": 4.563270777753791e-05,
|
| 253 |
+
"loss": 1.0385567474365234,
|
| 254 |
+
"step": 1400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.20286914939863787,
|
| 258 |
+
"eval_loss": 0.9979937076568604,
|
| 259 |
+
"eval_runtime": 193.9537,
|
| 260 |
+
"eval_samples_per_second": 119.853,
|
| 261 |
+
"eval_steps_per_second": 3.748,
|
| 262 |
+
"step": 1400
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.21011447616287496,
|
| 266 |
+
"grad_norm": 0.33952876925468445,
|
| 267 |
+
"learning_rate": 4.530117838210059e-05,
|
| 268 |
+
"loss": 1.040812759399414,
|
| 269 |
+
"step": 1450
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.21735980292711202,
|
| 273 |
+
"grad_norm": 0.3470812737941742,
|
| 274 |
+
"learning_rate": 4.4958819792379846e-05,
|
| 275 |
+
"loss": 1.0452130126953125,
|
| 276 |
+
"step": 1500
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.22460512969134908,
|
| 280 |
+
"grad_norm": 0.37180599570274353,
|
| 281 |
+
"learning_rate": 4.460581463165071e-05,
|
| 282 |
+
"loss": 1.0368424224853516,
|
| 283 |
+
"step": 1550
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.23185045645558613,
|
| 287 |
+
"grad_norm": 0.3716419041156769,
|
| 288 |
+
"learning_rate": 4.424235120235537e-05,
|
| 289 |
+
"loss": 1.0325569915771484,
|
| 290 |
+
"step": 1600
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.23185045645558613,
|
| 294 |
+
"eval_loss": 0.9863654971122742,
|
| 295 |
+
"eval_runtime": 192.694,
|
| 296 |
+
"eval_samples_per_second": 120.637,
|
| 297 |
+
"eval_steps_per_second": 3.773,
|
| 298 |
+
"step": 1600
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.23909578321982322,
|
| 302 |
+
"grad_norm": 0.33746030926704407,
|
| 303 |
+
"learning_rate": 4.386862338565759e-05,
|
| 304 |
+
"loss": 1.0336730194091797,
|
| 305 |
+
"step": 1650
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.24634110998406028,
|
| 309 |
+
"grad_norm": 0.34745824337005615,
|
| 310 |
+
"learning_rate": 4.3484830538021324e-05,
|
| 311 |
+
"loss": 1.0262679290771484,
|
| 312 |
+
"step": 1700
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 0.25358643674829734,
|
| 316 |
+
"grad_norm": 0.33224305510520935,
|
| 317 |
+
"learning_rate": 4.3091177384868585e-05,
|
| 318 |
+
"loss": 1.0238925170898439,
|
| 319 |
+
"step": 1750
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 0.2608317635125344,
|
| 323 |
+
"grad_norm": 0.3336823880672455,
|
| 324 |
+
"learning_rate": 4.26878739113734e-05,
|
| 325 |
+
"loss": 1.021321792602539,
|
| 326 |
+
"step": 1800
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.2608317635125344,
|
| 330 |
+
"eval_loss": 0.9752877354621887,
|
| 331 |
+
"eval_runtime": 192.2236,
|
| 332 |
+
"eval_samples_per_second": 120.932,
|
| 333 |
+
"eval_steps_per_second": 3.782,
|
| 334 |
+
"step": 1800
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 0.26807709027677146,
|
| 338 |
+
"grad_norm": 0.3285761773586273,
|
| 339 |
+
"learning_rate": 4.2275135250450106e-05,
|
| 340 |
+
"loss": 1.0217689514160155,
|
| 341 |
+
"step": 1850
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 0.27532241704100857,
|
| 345 |
+
"grad_norm": 0.35993409156799316,
|
| 346 |
+
"learning_rate": 4.1853181567995645e-05,
|
| 347 |
+
"loss": 1.0172651672363282,
|
| 348 |
+
"step": 1900
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 0.28256774380524563,
|
| 352 |
+
"grad_norm": 0.3507116734981537,
|
| 353 |
+
"learning_rate": 4.142223794544715e-05,
|
| 354 |
+
"loss": 1.0248129272460937,
|
| 355 |
+
"step": 1950
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.2898130705694827,
|
| 359 |
+
"grad_norm": 0.3568938374519348,
|
| 360 |
+
"learning_rate": 4.0982534259717475e-05,
|
| 361 |
+
"loss": 1.0136798095703126,
|
| 362 |
+
"step": 2000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.2898130705694827,
|
| 366 |
+
"eval_loss": 0.9648082256317139,
|
| 367 |
+
"eval_runtime": 192.1777,
|
| 368 |
+
"eval_samples_per_second": 120.961,
|
| 369 |
+
"eval_steps_per_second": 3.783,
|
| 370 |
+
"step": 2000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.29705839733371975,
|
| 374 |
+
"grad_norm": 0.3777235746383667,
|
| 375 |
+
"learning_rate": 4.053430506057268e-05,
|
| 376 |
+
"loss": 1.0077315521240235,
|
| 377 |
+
"step": 2050
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.3043037240979568,
|
| 381 |
+
"grad_norm": 0.3593011498451233,
|
| 382 |
+
"learning_rate": 4.0077789445516814e-05,
|
| 383 |
+
"loss": 1.0111034393310547,
|
| 384 |
+
"step": 2100
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.31154905086219387,
|
| 388 |
+
"grad_norm": 0.3624265193939209,
|
| 389 |
+
"learning_rate": 3.9613230932250985e-05,
|
| 390 |
+
"loss": 1.0038584136962891,
|
| 391 |
+
"step": 2150
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.3187943776264309,
|
| 395 |
+
"grad_norm": 0.3455412983894348,
|
| 396 |
+
"learning_rate": 3.9140877328774375e-05,
|
| 397 |
+
"loss": 1.0052964782714844,
|
| 398 |
+
"step": 2200
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.3187943776264309,
|
| 402 |
+
"eval_loss": 0.9546002149581909,
|
| 403 |
+
"eval_runtime": 193.0198,
|
| 404 |
+
"eval_samples_per_second": 120.433,
|
| 405 |
+
"eval_steps_per_second": 3.766,
|
| 406 |
+
"step": 2200
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.32603970439066804,
|
| 410 |
+
"grad_norm": 0.35707926750183105,
|
| 411 |
+
"learning_rate": 3.866098060119684e-05,
|
| 412 |
+
"loss": 0.9974722290039062,
|
| 413 |
+
"step": 2250
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.3332850311549051,
|
| 417 |
+
"grad_norm": 0.3994494080543518,
|
| 418 |
+
"learning_rate": 3.817379673933341e-05,
|
| 419 |
+
"loss": 0.9981327056884766,
|
| 420 |
+
"step": 2300
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.34053035791914216,
|
| 424 |
+
"grad_norm": 0.36747175455093384,
|
| 425 |
+
"learning_rate": 3.767958562015246e-05,
|
| 426 |
+
"loss": 0.9972872924804688,
|
| 427 |
+
"step": 2350
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.3477756846833792,
|
| 431 |
+
"grad_norm": 0.36322903633117676,
|
| 432 |
+
"learning_rate": 3.717861086915026e-05,
|
| 433 |
+
"loss": 0.9997342681884765,
|
| 434 |
+
"step": 2400
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.3477756846833792,
|
| 438 |
+
"eval_loss": 0.9470997452735901,
|
| 439 |
+
"eval_runtime": 192.2555,
|
| 440 |
+
"eval_samples_per_second": 120.912,
|
| 441 |
+
"eval_steps_per_second": 3.781,
|
| 442 |
+
"step": 2400
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.3550210114476163,
|
| 446 |
+
"grad_norm": 0.33153679966926575,
|
| 447 |
+
"learning_rate": 3.6671139719726174e-05,
|
| 448 |
+
"loss": 0.9987025451660156,
|
| 449 |
+
"step": 2450
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.36226633821185333,
|
| 453 |
+
"grad_norm": 0.40384331345558167,
|
| 454 |
+
"learning_rate": 3.6157442870633096e-05,
|
| 455 |
+
"loss": 0.992891616821289,
|
| 456 |
+
"step": 2500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.36951166497609045,
|
| 460 |
+
"grad_norm": 0.3685908913612366,
|
| 461 |
+
"learning_rate": 3.563779434157947e-05,
|
| 462 |
+
"loss": 0.991015396118164,
|
| 463 |
+
"step": 2550
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.3767569917403275,
|
| 467 |
+
"grad_norm": 0.37052083015441895,
|
| 468 |
+
"learning_rate": 3.511247132705986e-05,
|
| 469 |
+
"loss": 0.9884940338134766,
|
| 470 |
+
"step": 2600
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.3767569917403275,
|
| 474 |
+
"eval_loss": 0.939989447593689,
|
| 475 |
+
"eval_runtime": 192.4644,
|
| 476 |
+
"eval_samples_per_second": 120.781,
|
| 477 |
+
"eval_steps_per_second": 3.777,
|
| 478 |
+
"step": 2600
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.38400231850456457,
|
| 482 |
+
"grad_norm": 0.35503244400024414,
|
| 483 |
+
"learning_rate": 3.458175404849188e-05,
|
| 484 |
+
"loss": 0.9872270965576172,
|
| 485 |
+
"step": 2650
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3912476452688016,
|
| 489 |
+
"grad_norm": 0.3497767746448517,
|
| 490 |
+
"learning_rate": 3.4045925604738604e-05,
|
| 491 |
+
"loss": 0.9846116638183594,
|
| 492 |
+
"step": 2700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.3984929720330387,
|
| 496 |
+
"grad_norm": 0.3695673942565918,
|
| 497 |
+
"learning_rate": 3.350527182109603e-05,
|
| 498 |
+
"loss": 0.9809439849853515,
|
| 499 |
+
"step": 2750
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.40573829879727574,
|
| 503 |
+
"grad_norm": 0.36862626671791077,
|
| 504 |
+
"learning_rate": 3.296008109682616e-05,
|
| 505 |
+
"loss": 0.9826165008544921,
|
| 506 |
+
"step": 2800
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.40573829879727574,
|
| 510 |
+
"eval_loss": 0.9336075186729431,
|
| 511 |
+
"eval_runtime": 192.3219,
|
| 512 |
+
"eval_samples_per_second": 120.87,
|
| 513 |
+
"eval_steps_per_second": 3.78,
|
| 514 |
+
"step": 2800
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.4129836255615128,
|
| 518 |
+
"grad_norm": 0.34875786304473877,
|
| 519 |
+
"learning_rate": 3.241064425131708e-05,
|
| 520 |
+
"loss": 0.9786082458496094,
|
| 521 |
+
"step": 2850
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.4202289523257499,
|
| 525 |
+
"grad_norm": 0.3561500608921051,
|
| 526 |
+
"learning_rate": 3.185725436895209e-05,
|
| 527 |
+
"loss": 0.9792596435546875,
|
| 528 |
+
"step": 2900
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.427474279089987,
|
| 532 |
+
"grad_norm": 0.36314815282821655,
|
| 533 |
+
"learning_rate": 3.130020664277064e-05,
|
| 534 |
+
"loss": 0.9800941467285156,
|
| 535 |
+
"step": 2950
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.43471960585422403,
|
| 539 |
+
"grad_norm": 0.3698309361934662,
|
| 540 |
+
"learning_rate": 3.0739798217004354e-05,
|
| 541 |
+
"loss": 0.9847097015380859,
|
| 542 |
+
"step": 3000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.43471960585422403,
|
| 546 |
+
"eval_loss": 0.927081286907196,
|
| 547 |
+
"eval_runtime": 192.3461,
|
| 548 |
+
"eval_samples_per_second": 120.855,
|
| 549 |
+
"eval_steps_per_second": 3.78,
|
| 550 |
+
"step": 3000
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.4419649326184611,
|
| 554 |
+
"grad_norm": 0.3639983534812927,
|
| 555 |
+
"learning_rate": 3.0176328028572408e-05,
|
| 556 |
+
"loss": 0.9701497650146484,
|
| 557 |
+
"step": 3050
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 0.44921025938269815,
|
| 561 |
+
"grad_norm": 0.3515409827232361,
|
| 562 |
+
"learning_rate": 2.9610096647620445e-05,
|
| 563 |
+
"loss": 0.9721237945556641,
|
| 564 |
+
"step": 3100
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.4564555861469352,
|
| 568 |
+
"grad_norm": 0.391891747713089,
|
| 569 |
+
"learning_rate": 2.9041406117188458e-05,
|
| 570 |
+
"loss": 0.9730013275146484,
|
| 571 |
+
"step": 3150
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.46370091291117227,
|
| 575 |
+
"grad_norm": 0.3907790780067444,
|
| 576 |
+
"learning_rate": 2.8470559792092855e-05,
|
| 577 |
+
"loss": 0.974240951538086,
|
| 578 |
+
"step": 3200
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.46370091291117227,
|
| 582 |
+
"eval_loss": 0.9227471351623535,
|
| 583 |
+
"eval_runtime": 192.3251,
|
| 584 |
+
"eval_samples_per_second": 120.868,
|
| 585 |
+
"eval_steps_per_second": 3.78,
|
| 586 |
+
"step": 3200
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.4709462396754094,
|
| 590 |
+
"grad_norm": 0.3531961143016815,
|
| 591 |
+
"learning_rate": 2.789786217710888e-05,
|
| 592 |
+
"loss": 0.9673224639892578,
|
| 593 |
+
"step": 3250
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.47819156643964644,
|
| 597 |
+
"grad_norm": 0.33910030126571655,
|
| 598 |
+
"learning_rate": 2.732361876453957e-05,
|
| 599 |
+
"loss": 0.9719451904296875,
|
| 600 |
+
"step": 3300
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 0.4854368932038835,
|
| 604 |
+
"grad_norm": 0.37699881196022034,
|
| 605 |
+
"learning_rate": 2.6748135871257955e-05,
|
| 606 |
+
"loss": 0.9684635925292969,
|
| 607 |
+
"step": 3350
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 0.49268221996812056,
|
| 611 |
+
"grad_norm": 0.3599157929420471,
|
| 612 |
+
"learning_rate": 2.617172047530939e-05,
|
| 613 |
+
"loss": 0.957724609375,
|
| 614 |
+
"step": 3400
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.49268221996812056,
|
| 618 |
+
"eval_loss": 0.91666579246521,
|
| 619 |
+
"eval_runtime": 192.0058,
|
| 620 |
+
"eval_samples_per_second": 121.069,
|
| 621 |
+
"eval_steps_per_second": 3.786,
|
| 622 |
+
"step": 3400
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 0.4999275467323576,
|
| 626 |
+
"grad_norm": 0.37489375472068787,
|
| 627 |
+
"learning_rate": 2.5594680052161206e-05,
|
| 628 |
+
"loss": 0.9689096069335937,
|
| 629 |
+
"step": 3450
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 0.5071728734965947,
|
| 633 |
+
"grad_norm": 0.3328556418418884,
|
| 634 |
+
"learning_rate": 2.5017322410687077e-05,
|
| 635 |
+
"loss": 0.96274169921875,
|
| 636 |
+
"step": 3500
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 0.5144182002608317,
|
| 640 |
+
"grad_norm": 0.35683172941207886,
|
| 641 |
+
"learning_rate": 2.4439955528973414e-05,
|
| 642 |
+
"loss": 0.9614187622070313,
|
| 643 |
+
"step": 3550
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 0.5216635270250688,
|
| 647 |
+
"grad_norm": 0.36412957310676575,
|
| 648 |
+
"learning_rate": 2.386288739003567e-05,
|
| 649 |
+
"loss": 0.964275131225586,
|
| 650 |
+
"step": 3600
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.5216635270250688,
|
| 654 |
+
"eval_loss": 0.9119325280189514,
|
| 655 |
+
"eval_runtime": 192.5165,
|
| 656 |
+
"eval_samples_per_second": 120.748,
|
| 657 |
+
"eval_steps_per_second": 3.776,
|
| 658 |
+
"step": 3600
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.5289088537893059,
|
| 662 |
+
"grad_norm": 0.3563174307346344,
|
| 663 |
+
"learning_rate": 2.3286425817531836e-05,
|
| 664 |
+
"loss": 0.9626367950439453,
|
| 665 |
+
"step": 3650
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.5361541805535429,
|
| 669 |
+
"grad_norm": 0.3758934438228607,
|
| 670 |
+
"learning_rate": 2.271087831156107e-05,
|
| 671 |
+
"loss": 0.9564736938476562,
|
| 672 |
+
"step": 3700
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.5433995073177801,
|
| 676 |
+
"grad_norm": 0.34257784485816956,
|
| 677 |
+
"learning_rate": 2.2136551884634864e-05,
|
| 678 |
+
"loss": 0.9561862945556641,
|
| 679 |
+
"step": 3750
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 0.5506448340820171,
|
| 683 |
+
"grad_norm": 0.38688501715660095,
|
| 684 |
+
"learning_rate": 2.1563752897908352e-05,
|
| 685 |
+
"loss": 0.9572794342041016,
|
| 686 |
+
"step": 3800
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.5506448340820171,
|
| 690 |
+
"eval_loss": 0.9085790514945984,
|
| 691 |
+
"eval_runtime": 193.5054,
|
| 692 |
+
"eval_samples_per_second": 120.131,
|
| 693 |
+
"eval_steps_per_second": 3.757,
|
| 694 |
+
"step": 3800
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 0.5578901608462542,
|
| 698 |
+
"grad_norm": 0.4174833297729492,
|
| 699 |
+
"learning_rate": 2.0992786897758974e-05,
|
| 700 |
+
"loss": 0.9556180572509766,
|
| 701 |
+
"step": 3850
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 0.5651354876104913,
|
| 705 |
+
"grad_norm": 0.35218948125839233,
|
| 706 |
+
"learning_rate": 2.0423958452799854e-05,
|
| 707 |
+
"loss": 0.9547309875488281,
|
| 708 |
+
"step": 3900
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.5723808143747283,
|
| 712 |
+
"grad_norm": 0.3613223433494568,
|
| 713 |
+
"learning_rate": 1.9857570991414737e-05,
|
| 714 |
+
"loss": 0.9548380279541016,
|
| 715 |
+
"step": 3950
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 0.5796261411389654,
|
| 719 |
+
"grad_norm": 0.35305657982826233,
|
| 720 |
+
"learning_rate": 1.9293926639901092e-05,
|
| 721 |
+
"loss": 0.9536138916015625,
|
| 722 |
+
"step": 4000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.5796261411389654,
|
| 726 |
+
"eval_loss": 0.9045887589454651,
|
| 727 |
+
"eval_runtime": 193.4058,
|
| 728 |
+
"eval_samples_per_second": 120.193,
|
| 729 |
+
"eval_steps_per_second": 3.759,
|
| 730 |
+
"step": 4000
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.5868714679032024,
|
| 734 |
+
"grad_norm": 0.35594019293785095,
|
| 735 |
+
"learning_rate": 1.873332606130787e-05,
|
| 736 |
+
"loss": 0.961040267944336,
|
| 737 |
+
"step": 4050
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5941167946674395,
|
| 741 |
+
"grad_norm": 0.3579290509223938,
|
| 742 |
+
"learning_rate": 1.8176068295053684e-05,
|
| 743 |
+
"loss": 0.953524169921875,
|
| 744 |
+
"step": 4100
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.6013621214316766,
|
| 748 |
+
"grad_norm": 0.37046733498573303,
|
| 749 |
+
"learning_rate": 1.7622450597411216e-05,
|
| 750 |
+
"loss": 0.9559716796875,
|
| 751 |
+
"step": 4150
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.6086074481959136,
|
| 755 |
+
"grad_norm": 0.41828927397727966,
|
| 756 |
+
"learning_rate": 1.7072768282942695e-05,
|
| 757 |
+
"loss": 0.9484828186035156,
|
| 758 |
+
"step": 4200
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.6086074481959136,
|
| 762 |
+
"eval_loss": 0.9021432995796204,
|
| 763 |
+
"eval_runtime": 193.4682,
|
| 764 |
+
"eval_samples_per_second": 120.154,
|
| 765 |
+
"eval_steps_per_second": 3.758,
|
| 766 |
+
"step": 4200
|
| 767 |
+
}
|
| 768 |
+
],
|
| 769 |
+
"logging_steps": 50,
|
| 770 |
+
"max_steps": 6901,
|
| 771 |
+
"num_input_tokens_seen": 0,
|
| 772 |
+
"num_train_epochs": 1,
|
| 773 |
+
"save_steps": 200,
|
| 774 |
+
"stateful_callbacks": {
|
| 775 |
+
"TrainerControl": {
|
| 776 |
+
"args": {
|
| 777 |
+
"should_epoch_stop": false,
|
| 778 |
+
"should_evaluate": false,
|
| 779 |
+
"should_log": false,
|
| 780 |
+
"should_save": true,
|
| 781 |
+
"should_training_stop": false
|
| 782 |
+
},
|
| 783 |
+
"attributes": {}
|
| 784 |
+
}
|
| 785 |
+
},
|
| 786 |
+
"total_flos": 8789181767221248.0,
|
| 787 |
+
"train_batch_size": 32,
|
| 788 |
+
"trial_name": null,
|
| 789 |
+
"trial_params": null
|
| 790 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11eeaf5563c5febc84d935fa7fbde88371bd4c9b6600b553f7c114370c117c76
|
| 3 |
+
size 5137
|