c3
Browse files- config.json +70 -0
- generation_config.json +7 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer_config.json +20 -0
- trainer_state.json +651 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2",
|
| 3 |
+
"activation_function": "leaky_relu",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_block_resid_gain": 1,
|
| 8 |
+
"attn_block_skip_gain": 1,
|
| 9 |
+
"attn_mat_resid_gain": 1,
|
| 10 |
+
"attn_mat_skip_gain": 0,
|
| 11 |
+
"attn_pdrop": 0,
|
| 12 |
+
"bos_token_id": 0,
|
| 13 |
+
"centre_attn": false,
|
| 14 |
+
"centre_attn_gain": 1.0,
|
| 15 |
+
"embd_pdrop": 0,
|
| 16 |
+
"eos_token_id": 0,
|
| 17 |
+
"first_layer_value_resid_gain": null,
|
| 18 |
+
"initializer_range": 0.02,
|
| 19 |
+
"key_init_std": null,
|
| 20 |
+
"last_layer_proj_resid_gain": null,
|
| 21 |
+
"layer_norm_epsilon": 1e-05,
|
| 22 |
+
"lrelu_neg_slope": 0,
|
| 23 |
+
"mlp_block_resid_gain": 1,
|
| 24 |
+
"mlp_block_skip_gain": 1,
|
| 25 |
+
"mlp_proj_init_std": false,
|
| 26 |
+
"model_type": "gpt2",
|
| 27 |
+
"n_ctx": 128,
|
| 28 |
+
"n_embd": 117,
|
| 29 |
+
"n_head": 9,
|
| 30 |
+
"n_inner": 468,
|
| 31 |
+
"n_layer": 12,
|
| 32 |
+
"n_positions": 1024,
|
| 33 |
+
"norm_position": "pre",
|
| 34 |
+
"norm_type": "rmsnorm",
|
| 35 |
+
"output_attentions": "false",
|
| 36 |
+
"parallel_layers": false,
|
| 37 |
+
"proj_init_type": "normal",
|
| 38 |
+
"proj_resid_gain": 1.0,
|
| 39 |
+
"proj_skip_gain": null,
|
| 40 |
+
"query_init_std": null,
|
| 41 |
+
"reorder_and_upcast_attn": false,
|
| 42 |
+
"resid_pdrop": 0,
|
| 43 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 44 |
+
"scale_attn_weights": true,
|
| 45 |
+
"summary_activation": null,
|
| 46 |
+
"summary_first_dropout": 0.1,
|
| 47 |
+
"summary_proj_to_labels": true,
|
| 48 |
+
"summary_type": "cls_index",
|
| 49 |
+
"summary_use_proj": true,
|
| 50 |
+
"task_specific_params": {
|
| 51 |
+
"text-generation": {
|
| 52 |
+
"do_sample": true,
|
| 53 |
+
"max_length": 50
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"tie_valproj_init": null,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"trainable_attn_block_gains": false,
|
| 59 |
+
"trainable_attn_mat_gains": false,
|
| 60 |
+
"trainable_mlp_block_gains": false,
|
| 61 |
+
"trainable_proj_gains": false,
|
| 62 |
+
"trainable_value_gains": false,
|
| 63 |
+
"transformers_version": "4.38.1",
|
| 64 |
+
"use_cache": true,
|
| 65 |
+
"val_init_type": "normal",
|
| 66 |
+
"val_proj_init_std": null,
|
| 67 |
+
"value_resid_gain": 1,
|
| 68 |
+
"value_skip_gain": 0,
|
| 69 |
+
"vocab_size": 50000
|
| 70 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"output_attentions": "false",
|
| 6 |
+
"transformers_version": "4.38.1"
|
| 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:a9ae76f9027e186f775f3e7b2ec3d8e7d932aec4db025f0b3bf4cc3ee2d9bc09
|
| 3 |
+
size 31854136
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c1e3b3ed8755dc8255367b3729206f73c8a274cd36c62868d1a1951b6c2a6773
|
| 3 |
+
size 63654074
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51fbb656db90359360bf0f30e9bfe6384141ed3c5d5b9e95ccb660b0f4a33dea
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ef14ddcfd22ab488bed70e198ffa875ca0f31ed0bacd8dca3390eeaf6c7fd9d
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|endoftext|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<|endoftext|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
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 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
}
|
| 12 |
+
},
|
| 13 |
+
"bos_token": "<|endoftext|>",
|
| 14 |
+
"clean_up_tokenization_spaces": true,
|
| 15 |
+
"eos_token": "<|endoftext|>",
|
| 16 |
+
"model_max_length": 1024,
|
| 17 |
+
"pad_token": "<|endoftext|>",
|
| 18 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 19 |
+
"unk_token": "<|endoftext|>"
|
| 20 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,651 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9795004547680682,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 21000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 1.301074743270874,
|
| 14 |
+
"learning_rate": 0.00046641791044776124,
|
| 15 |
+
"loss": 7.611,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"eval_loss": 5.238980293273926,
|
| 21 |
+
"eval_runtime": 567.2757,
|
| 22 |
+
"eval_samples_per_second": 99.477,
|
| 23 |
+
"eval_steps_per_second": 3.11,
|
| 24 |
+
"step": 500
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.05,
|
| 28 |
+
"grad_norm": 1.1419352293014526,
|
| 29 |
+
"learning_rate": 0.0009328358208955225,
|
| 30 |
+
"loss": 4.4792,
|
| 31 |
+
"step": 1000
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.05,
|
| 35 |
+
"eval_loss": 4.0363240242004395,
|
| 36 |
+
"eval_runtime": 568.283,
|
| 37 |
+
"eval_samples_per_second": 99.301,
|
| 38 |
+
"eval_steps_per_second": 3.104,
|
| 39 |
+
"step": 1000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.07,
|
| 43 |
+
"grad_norm": 1.1277365684509277,
|
| 44 |
+
"learning_rate": 0.0009789877154220063,
|
| 45 |
+
"loss": 3.6503,
|
| 46 |
+
"step": 1500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 0.07,
|
| 50 |
+
"eval_loss": 3.468844175338745,
|
| 51 |
+
"eval_runtime": 568.7397,
|
| 52 |
+
"eval_samples_per_second": 99.221,
|
| 53 |
+
"eval_steps_per_second": 3.102,
|
| 54 |
+
"step": 1500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.09,
|
| 58 |
+
"grad_norm": 1.1394764184951782,
|
| 59 |
+
"learning_rate": 0.0009544406539991162,
|
| 60 |
+
"loss": 3.2321,
|
| 61 |
+
"step": 2000
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.09,
|
| 65 |
+
"eval_loss": 3.1853878498077393,
|
| 66 |
+
"eval_runtime": 568.7397,
|
| 67 |
+
"eval_samples_per_second": 99.221,
|
| 68 |
+
"eval_steps_per_second": 3.102,
|
| 69 |
+
"step": 2000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.12,
|
| 73 |
+
"grad_norm": 1.0943962335586548,
|
| 74 |
+
"learning_rate": 0.0009298935925762261,
|
| 75 |
+
"loss": 3.0165,
|
| 76 |
+
"step": 2500
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.12,
|
| 80 |
+
"eval_loss": 3.0076937675476074,
|
| 81 |
+
"eval_runtime": 569.7751,
|
| 82 |
+
"eval_samples_per_second": 99.041,
|
| 83 |
+
"eval_steps_per_second": 3.096,
|
| 84 |
+
"step": 2500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.14,
|
| 88 |
+
"grad_norm": 0.9250391721725464,
|
| 89 |
+
"learning_rate": 0.0009053465311533362,
|
| 90 |
+
"loss": 2.8757,
|
| 91 |
+
"step": 3000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.14,
|
| 95 |
+
"eval_loss": 2.8989241123199463,
|
| 96 |
+
"eval_runtime": 569.8016,
|
| 97 |
+
"eval_samples_per_second": 99.036,
|
| 98 |
+
"eval_steps_per_second": 3.096,
|
| 99 |
+
"step": 3000
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.16,
|
| 103 |
+
"grad_norm": 0.9219993948936462,
|
| 104 |
+
"learning_rate": 0.0008807994697304462,
|
| 105 |
+
"loss": 2.7831,
|
| 106 |
+
"step": 3500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.16,
|
| 110 |
+
"eval_loss": 2.815277099609375,
|
| 111 |
+
"eval_runtime": 569.7431,
|
| 112 |
+
"eval_samples_per_second": 99.046,
|
| 113 |
+
"eval_steps_per_second": 3.096,
|
| 114 |
+
"step": 3500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.19,
|
| 118 |
+
"grad_norm": 1.0010333061218262,
|
| 119 |
+
"learning_rate": 0.0008562524083075562,
|
| 120 |
+
"loss": 2.7098,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.19,
|
| 125 |
+
"eval_loss": 2.7535383701324463,
|
| 126 |
+
"eval_runtime": 564.7218,
|
| 127 |
+
"eval_samples_per_second": 99.927,
|
| 128 |
+
"eval_steps_per_second": 3.124,
|
| 129 |
+
"step": 4000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.21,
|
| 133 |
+
"grad_norm": 0.9716792702674866,
|
| 134 |
+
"learning_rate": 0.0008317053468846664,
|
| 135 |
+
"loss": 2.6551,
|
| 136 |
+
"step": 4500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.21,
|
| 140 |
+
"eval_loss": 2.7090485095977783,
|
| 141 |
+
"eval_runtime": 566.8661,
|
| 142 |
+
"eval_samples_per_second": 99.549,
|
| 143 |
+
"eval_steps_per_second": 3.112,
|
| 144 |
+
"step": 4500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.23,
|
| 148 |
+
"grad_norm": 0.9213451743125916,
|
| 149 |
+
"learning_rate": 0.0008071582854617764,
|
| 150 |
+
"loss": 2.6152,
|
| 151 |
+
"step": 5000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.23,
|
| 155 |
+
"eval_loss": 2.666491746902466,
|
| 156 |
+
"eval_runtime": 565.311,
|
| 157 |
+
"eval_samples_per_second": 99.823,
|
| 158 |
+
"eval_steps_per_second": 3.12,
|
| 159 |
+
"step": 5000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.26,
|
| 163 |
+
"grad_norm": 0.8985408544540405,
|
| 164 |
+
"learning_rate": 0.0007826112240388863,
|
| 165 |
+
"loss": 2.575,
|
| 166 |
+
"step": 5500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.26,
|
| 170 |
+
"eval_loss": 2.6350765228271484,
|
| 171 |
+
"eval_runtime": 564.212,
|
| 172 |
+
"eval_samples_per_second": 100.017,
|
| 173 |
+
"eval_steps_per_second": 3.126,
|
| 174 |
+
"step": 5500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.28,
|
| 178 |
+
"grad_norm": 0.8632990717887878,
|
| 179 |
+
"learning_rate": 0.0007580641626159963,
|
| 180 |
+
"loss": 2.5464,
|
| 181 |
+
"step": 6000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.28,
|
| 185 |
+
"eval_loss": 2.6097421646118164,
|
| 186 |
+
"eval_runtime": 568.485,
|
| 187 |
+
"eval_samples_per_second": 99.266,
|
| 188 |
+
"eval_steps_per_second": 3.103,
|
| 189 |
+
"step": 6000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.3,
|
| 193 |
+
"grad_norm": 0.8819186687469482,
|
| 194 |
+
"learning_rate": 0.0007335171011931065,
|
| 195 |
+
"loss": 2.5183,
|
| 196 |
+
"step": 6500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.3,
|
| 200 |
+
"eval_loss": 2.580737590789795,
|
| 201 |
+
"eval_runtime": 568.2279,
|
| 202 |
+
"eval_samples_per_second": 99.311,
|
| 203 |
+
"eval_steps_per_second": 3.104,
|
| 204 |
+
"step": 6500
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.33,
|
| 208 |
+
"grad_norm": 0.9012317657470703,
|
| 209 |
+
"learning_rate": 0.0007089700397702165,
|
| 210 |
+
"loss": 2.4977,
|
| 211 |
+
"step": 7000
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.33,
|
| 215 |
+
"eval_loss": 2.5541210174560547,
|
| 216 |
+
"eval_runtime": 568.3683,
|
| 217 |
+
"eval_samples_per_second": 99.286,
|
| 218 |
+
"eval_steps_per_second": 3.104,
|
| 219 |
+
"step": 7000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.35,
|
| 223 |
+
"grad_norm": 0.9294862151145935,
|
| 224 |
+
"learning_rate": 0.0006844229783473265,
|
| 225 |
+
"loss": 2.4723,
|
| 226 |
+
"step": 7500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.35,
|
| 230 |
+
"eval_loss": 2.535729169845581,
|
| 231 |
+
"eval_runtime": 569.135,
|
| 232 |
+
"eval_samples_per_second": 99.152,
|
| 233 |
+
"eval_steps_per_second": 3.099,
|
| 234 |
+
"step": 7500
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.37,
|
| 238 |
+
"grad_norm": 0.9035699367523193,
|
| 239 |
+
"learning_rate": 0.0006598759169244364,
|
| 240 |
+
"loss": 2.4549,
|
| 241 |
+
"step": 8000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.37,
|
| 245 |
+
"eval_loss": 2.5194778442382812,
|
| 246 |
+
"eval_runtime": 569.6481,
|
| 247 |
+
"eval_samples_per_second": 99.063,
|
| 248 |
+
"eval_steps_per_second": 3.097,
|
| 249 |
+
"step": 8000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.4,
|
| 253 |
+
"grad_norm": 0.9459491968154907,
|
| 254 |
+
"learning_rate": 0.0006353779496243923,
|
| 255 |
+
"loss": 2.4428,
|
| 256 |
+
"step": 8500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.4,
|
| 260 |
+
"eval_loss": 2.502213716506958,
|
| 261 |
+
"eval_runtime": 565.3877,
|
| 262 |
+
"eval_samples_per_second": 99.809,
|
| 263 |
+
"eval_steps_per_second": 3.12,
|
| 264 |
+
"step": 8500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.42,
|
| 268 |
+
"grad_norm": 0.8880784511566162,
|
| 269 |
+
"learning_rate": 0.0006108308882015024,
|
| 270 |
+
"loss": 2.4255,
|
| 271 |
+
"step": 9000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.42,
|
| 275 |
+
"eval_loss": 2.4834067821502686,
|
| 276 |
+
"eval_runtime": 567.6005,
|
| 277 |
+
"eval_samples_per_second": 99.42,
|
| 278 |
+
"eval_steps_per_second": 3.108,
|
| 279 |
+
"step": 9000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.44,
|
| 283 |
+
"grad_norm": 0.9531754851341248,
|
| 284 |
+
"learning_rate": 0.0005862838267786123,
|
| 285 |
+
"loss": 2.4092,
|
| 286 |
+
"step": 9500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.44,
|
| 290 |
+
"eval_loss": 2.473543643951416,
|
| 291 |
+
"eval_runtime": 561.7685,
|
| 292 |
+
"eval_samples_per_second": 100.452,
|
| 293 |
+
"eval_steps_per_second": 3.14,
|
| 294 |
+
"step": 9500
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.47,
|
| 298 |
+
"grad_norm": 0.9443451762199402,
|
| 299 |
+
"learning_rate": 0.0005617367653557224,
|
| 300 |
+
"loss": 2.3859,
|
| 301 |
+
"step": 10000
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 0.47,
|
| 305 |
+
"eval_loss": 2.456810712814331,
|
| 306 |
+
"eval_runtime": 563.5685,
|
| 307 |
+
"eval_samples_per_second": 100.132,
|
| 308 |
+
"eval_steps_per_second": 3.13,
|
| 309 |
+
"step": 10000
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.49,
|
| 313 |
+
"grad_norm": 0.9646140933036804,
|
| 314 |
+
"learning_rate": 0.0005372387980556782,
|
| 315 |
+
"loss": 2.3753,
|
| 316 |
+
"step": 10500
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.49,
|
| 320 |
+
"eval_loss": 2.44191837310791,
|
| 321 |
+
"eval_runtime": 563.9978,
|
| 322 |
+
"eval_samples_per_second": 100.055,
|
| 323 |
+
"eval_steps_per_second": 3.128,
|
| 324 |
+
"step": 10500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.51,
|
| 328 |
+
"grad_norm": 0.9498624801635742,
|
| 329 |
+
"learning_rate": 0.0005126917366327883,
|
| 330 |
+
"loss": 2.3625,
|
| 331 |
+
"step": 11000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.51,
|
| 335 |
+
"eval_loss": 2.428806781768799,
|
| 336 |
+
"eval_runtime": 563.9618,
|
| 337 |
+
"eval_samples_per_second": 100.062,
|
| 338 |
+
"eval_steps_per_second": 3.128,
|
| 339 |
+
"step": 11000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.54,
|
| 343 |
+
"grad_norm": 0.9920567870140076,
|
| 344 |
+
"learning_rate": 0.0004881446752098984,
|
| 345 |
+
"loss": 2.3547,
|
| 346 |
+
"step": 11500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.54,
|
| 350 |
+
"eval_loss": 2.4165408611297607,
|
| 351 |
+
"eval_runtime": 563.8141,
|
| 352 |
+
"eval_samples_per_second": 100.088,
|
| 353 |
+
"eval_steps_per_second": 3.129,
|
| 354 |
+
"step": 11500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.56,
|
| 358 |
+
"grad_norm": 0.9328219294548035,
|
| 359 |
+
"learning_rate": 0.0004635976137870084,
|
| 360 |
+
"loss": 2.3387,
|
| 361 |
+
"step": 12000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.56,
|
| 365 |
+
"eval_loss": 2.40574312210083,
|
| 366 |
+
"eval_runtime": 563.9451,
|
| 367 |
+
"eval_samples_per_second": 100.065,
|
| 368 |
+
"eval_steps_per_second": 3.128,
|
| 369 |
+
"step": 12000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.58,
|
| 373 |
+
"grad_norm": 0.9478664994239807,
|
| 374 |
+
"learning_rate": 0.00043909964648696415,
|
| 375 |
+
"loss": 2.3277,
|
| 376 |
+
"step": 12500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 0.58,
|
| 380 |
+
"eval_loss": 2.390106201171875,
|
| 381 |
+
"eval_runtime": 563.4302,
|
| 382 |
+
"eval_samples_per_second": 100.156,
|
| 383 |
+
"eval_steps_per_second": 3.131,
|
| 384 |
+
"step": 12500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.61,
|
| 388 |
+
"grad_norm": 0.9822357296943665,
|
| 389 |
+
"learning_rate": 0.0004145525850640742,
|
| 390 |
+
"loss": 2.3071,
|
| 391 |
+
"step": 13000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.61,
|
| 395 |
+
"eval_loss": 2.380479335784912,
|
| 396 |
+
"eval_runtime": 563.2217,
|
| 397 |
+
"eval_samples_per_second": 100.193,
|
| 398 |
+
"eval_steps_per_second": 3.132,
|
| 399 |
+
"step": 13000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.63,
|
| 403 |
+
"grad_norm": 0.9824458956718445,
|
| 404 |
+
"learning_rate": 0.00039000552364118425,
|
| 405 |
+
"loss": 2.3087,
|
| 406 |
+
"step": 13500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.63,
|
| 410 |
+
"eval_loss": 2.366305112838745,
|
| 411 |
+
"eval_runtime": 563.1953,
|
| 412 |
+
"eval_samples_per_second": 100.198,
|
| 413 |
+
"eval_steps_per_second": 3.132,
|
| 414 |
+
"step": 13500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.65,
|
| 418 |
+
"grad_norm": 1.0211139917373657,
|
| 419 |
+
"learning_rate": 0.00036545846221829433,
|
| 420 |
+
"loss": 2.283,
|
| 421 |
+
"step": 14000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.65,
|
| 425 |
+
"eval_loss": 2.3575291633605957,
|
| 426 |
+
"eval_runtime": 563.2425,
|
| 427 |
+
"eval_samples_per_second": 100.19,
|
| 428 |
+
"eval_steps_per_second": 3.132,
|
| 429 |
+
"step": 14000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.68,
|
| 433 |
+
"grad_norm": 1.0012933015823364,
|
| 434 |
+
"learning_rate": 0.00034091140079540435,
|
| 435 |
+
"loss": 2.2716,
|
| 436 |
+
"step": 14500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.68,
|
| 440 |
+
"eval_loss": 2.3457164764404297,
|
| 441 |
+
"eval_runtime": 562.9008,
|
| 442 |
+
"eval_samples_per_second": 100.25,
|
| 443 |
+
"eval_steps_per_second": 3.134,
|
| 444 |
+
"step": 14500
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.7,
|
| 448 |
+
"grad_norm": 1.0424264669418335,
|
| 449 |
+
"learning_rate": 0.0003164134334953601,
|
| 450 |
+
"loss": 2.2709,
|
| 451 |
+
"step": 15000
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.7,
|
| 455 |
+
"eval_loss": 2.3343236446380615,
|
| 456 |
+
"eval_runtime": 562.6281,
|
| 457 |
+
"eval_samples_per_second": 100.299,
|
| 458 |
+
"eval_steps_per_second": 3.135,
|
| 459 |
+
"step": 15000
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.72,
|
| 463 |
+
"grad_norm": 1.1129311323165894,
|
| 464 |
+
"learning_rate": 0.00029186637207247017,
|
| 465 |
+
"loss": 2.2584,
|
| 466 |
+
"step": 15500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.72,
|
| 470 |
+
"eval_loss": 2.3277084827423096,
|
| 471 |
+
"eval_runtime": 574.0103,
|
| 472 |
+
"eval_samples_per_second": 98.31,
|
| 473 |
+
"eval_steps_per_second": 3.073,
|
| 474 |
+
"step": 15500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.75,
|
| 478 |
+
"grad_norm": 1.0602174997329712,
|
| 479 |
+
"learning_rate": 0.0002673193106495802,
|
| 480 |
+
"loss": 2.2509,
|
| 481 |
+
"step": 16000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.75,
|
| 485 |
+
"eval_loss": 2.3144421577453613,
|
| 486 |
+
"eval_runtime": 566.0389,
|
| 487 |
+
"eval_samples_per_second": 99.695,
|
| 488 |
+
"eval_steps_per_second": 3.116,
|
| 489 |
+
"step": 16000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.77,
|
| 493 |
+
"grad_norm": 1.1678231954574585,
|
| 494 |
+
"learning_rate": 0.00024277224922669027,
|
| 495 |
+
"loss": 2.237,
|
| 496 |
+
"step": 16500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.77,
|
| 500 |
+
"eval_loss": 2.305784225463867,
|
| 501 |
+
"eval_runtime": 563.0281,
|
| 502 |
+
"eval_samples_per_second": 100.228,
|
| 503 |
+
"eval_steps_per_second": 3.133,
|
| 504 |
+
"step": 16500
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.79,
|
| 508 |
+
"grad_norm": 1.1629687547683716,
|
| 509 |
+
"learning_rate": 0.000218274281926646,
|
| 510 |
+
"loss": 2.2249,
|
| 511 |
+
"step": 17000
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.79,
|
| 515 |
+
"eval_loss": 2.2936198711395264,
|
| 516 |
+
"eval_runtime": 562.9504,
|
| 517 |
+
"eval_samples_per_second": 100.242,
|
| 518 |
+
"eval_steps_per_second": 3.133,
|
| 519 |
+
"step": 17000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.82,
|
| 523 |
+
"grad_norm": 1.0790985822677612,
|
| 524 |
+
"learning_rate": 0.00019372722050375605,
|
| 525 |
+
"loss": 2.219,
|
| 526 |
+
"step": 17500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.82,
|
| 530 |
+
"eval_loss": 2.2830076217651367,
|
| 531 |
+
"eval_runtime": 562.6154,
|
| 532 |
+
"eval_samples_per_second": 100.301,
|
| 533 |
+
"eval_steps_per_second": 3.135,
|
| 534 |
+
"step": 17500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.84,
|
| 538 |
+
"grad_norm": 1.157651424407959,
|
| 539 |
+
"learning_rate": 0.0001691801590808661,
|
| 540 |
+
"loss": 2.1999,
|
| 541 |
+
"step": 18000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.84,
|
| 545 |
+
"eval_loss": 2.2722585201263428,
|
| 546 |
+
"eval_runtime": 562.9469,
|
| 547 |
+
"eval_samples_per_second": 100.242,
|
| 548 |
+
"eval_steps_per_second": 3.134,
|
| 549 |
+
"step": 18000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.86,
|
| 553 |
+
"grad_norm": 1.0704936981201172,
|
| 554 |
+
"learning_rate": 0.00014463309765797616,
|
| 555 |
+
"loss": 2.1999,
|
| 556 |
+
"step": 18500
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.86,
|
| 560 |
+
"eval_loss": 2.2624754905700684,
|
| 561 |
+
"eval_runtime": 562.9358,
|
| 562 |
+
"eval_samples_per_second": 100.244,
|
| 563 |
+
"eval_steps_per_second": 3.134,
|
| 564 |
+
"step": 18500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.89,
|
| 568 |
+
"grad_norm": 1.1210389137268066,
|
| 569 |
+
"learning_rate": 0.00012008603623508621,
|
| 570 |
+
"loss": 2.1898,
|
| 571 |
+
"step": 19000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 0.89,
|
| 575 |
+
"eval_loss": 2.254652976989746,
|
| 576 |
+
"eval_runtime": 563.3356,
|
| 577 |
+
"eval_samples_per_second": 100.173,
|
| 578 |
+
"eval_steps_per_second": 3.131,
|
| 579 |
+
"step": 19000
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.91,
|
| 583 |
+
"grad_norm": 1.0510447025299072,
|
| 584 |
+
"learning_rate": 9.558806893504194e-05,
|
| 585 |
+
"loss": 2.1796,
|
| 586 |
+
"step": 19500
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.91,
|
| 590 |
+
"eval_loss": 2.2447702884674072,
|
| 591 |
+
"eval_runtime": 563.3055,
|
| 592 |
+
"eval_samples_per_second": 100.178,
|
| 593 |
+
"eval_steps_per_second": 3.132,
|
| 594 |
+
"step": 19500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.93,
|
| 598 |
+
"grad_norm": 1.0447243452072144,
|
| 599 |
+
"learning_rate": 7.1041007512152e-05,
|
| 600 |
+
"loss": 2.1649,
|
| 601 |
+
"step": 20000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.93,
|
| 605 |
+
"eval_loss": 2.236150026321411,
|
| 606 |
+
"eval_runtime": 563.3424,
|
| 607 |
+
"eval_samples_per_second": 100.172,
|
| 608 |
+
"eval_steps_per_second": 3.131,
|
| 609 |
+
"step": 20000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.96,
|
| 613 |
+
"grad_norm": 1.039281964302063,
|
| 614 |
+
"learning_rate": 4.649394608926206e-05,
|
| 615 |
+
"loss": 2.1606,
|
| 616 |
+
"step": 20500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 0.96,
|
| 620 |
+
"eval_loss": 2.228179693222046,
|
| 621 |
+
"eval_runtime": 563.3883,
|
| 622 |
+
"eval_samples_per_second": 100.164,
|
| 623 |
+
"eval_steps_per_second": 3.131,
|
| 624 |
+
"step": 20500
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.98,
|
| 628 |
+
"grad_norm": 1.0719681978225708,
|
| 629 |
+
"learning_rate": 2.1946884666372103e-05,
|
| 630 |
+
"loss": 2.1521,
|
| 631 |
+
"step": 21000
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.98,
|
| 635 |
+
"eval_loss": 2.221137285232544,
|
| 636 |
+
"eval_runtime": 563.1866,
|
| 637 |
+
"eval_samples_per_second": 100.199,
|
| 638 |
+
"eval_steps_per_second": 3.132,
|
| 639 |
+
"step": 21000
|
| 640 |
+
}
|
| 641 |
+
],
|
| 642 |
+
"logging_steps": 500,
|
| 643 |
+
"max_steps": 21439,
|
| 644 |
+
"num_input_tokens_seen": 0,
|
| 645 |
+
"num_train_epochs": 1,
|
| 646 |
+
"save_steps": 7000,
|
| 647 |
+
"total_flos": 4102529679360000.0,
|
| 648 |
+
"train_batch_size": 32,
|
| 649 |
+
"trial_name": null,
|
| 650 |
+
"trial_params": null
|
| 651 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:546837c0781daaeabb91a4ad5c2d28d1b83c6d2c1b686cbf09ee4bf60e2380b5
|
| 3 |
+
size 4920
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|