commit from $USER
Browse files- config.json +39 -0
- generation_config.json +6 -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 +30 -0
- tokenizer_config.json +22 -0
- trainer_state.json +1036 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./gpt2-shakespeare\\checkpoint-3900",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": false,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.40.1",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50257
|
| 39 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.40.1"
|
| 6 |
+
}
|
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 sha256:5154c6ca38077b04d8f4173c74b2da3d46a04580fc6c8fe1432a08b733160735
|
| 3 |
+
size 497774208
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36bc7b32830f2fadb2a016745d520bddc6aecee7360385d143eeaf0c65cbff20
|
| 3 |
+
size 995638202
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61c89b91383aabdf3598cc50da4f10b585def58e826bb18972264e57b5b9ca5d
|
| 3 |
+
size 13990
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa53da977e9f71f807e5e42d8f652a9a84db1910f5ed30f09f7564220a0a6e4c
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": {
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
}
|
| 13 |
+
},
|
| 14 |
+
"bos_token": "<|endoftext|>",
|
| 15 |
+
"clean_up_tokenization_spaces": true,
|
| 16 |
+
"eos_token": "<|endoftext|>",
|
| 17 |
+
"errors": "replace",
|
| 18 |
+
"model_max_length": 1024,
|
| 19 |
+
"pad_token": "<|endoftext|>",
|
| 20 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 21 |
+
"unk_token": "<|endoftext|>"
|
| 22 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1036 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.3169845594913716,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 4350,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.009082652134423252,
|
| 13 |
+
"grad_norm": 2.5449585914611816,
|
| 14 |
+
"learning_rate": 4.9848622464426284e-05,
|
| 15 |
+
"loss": 8.7191,
|
| 16 |
+
"step": 30
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.018165304268846504,
|
| 20 |
+
"grad_norm": 2.3144371509552,
|
| 21 |
+
"learning_rate": 4.969724492885256e-05,
|
| 22 |
+
"loss": 7.4698,
|
| 23 |
+
"step": 60
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.027247956403269755,
|
| 27 |
+
"grad_norm": 2.304499626159668,
|
| 28 |
+
"learning_rate": 4.954586739327884e-05,
|
| 29 |
+
"loss": 6.5589,
|
| 30 |
+
"step": 90
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.03633060853769301,
|
| 34 |
+
"grad_norm": 2.272608757019043,
|
| 35 |
+
"learning_rate": 4.9394489857705115e-05,
|
| 36 |
+
"loss": 6.2425,
|
| 37 |
+
"step": 120
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.045413260672116255,
|
| 41 |
+
"grad_norm": 2.46329402923584,
|
| 42 |
+
"learning_rate": 4.9243112322131396e-05,
|
| 43 |
+
"loss": 6.1459,
|
| 44 |
+
"step": 150
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05449591280653951,
|
| 48 |
+
"grad_norm": 1.8283530473709106,
|
| 49 |
+
"learning_rate": 4.909173478655768e-05,
|
| 50 |
+
"loss": 5.969,
|
| 51 |
+
"step": 180
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.06357856494096276,
|
| 55 |
+
"grad_norm": 2.1723110675811768,
|
| 56 |
+
"learning_rate": 4.894035725098395e-05,
|
| 57 |
+
"loss": 6.008,
|
| 58 |
+
"step": 210
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.07266121707538602,
|
| 62 |
+
"grad_norm": 2.5368807315826416,
|
| 63 |
+
"learning_rate": 4.878897971541024e-05,
|
| 64 |
+
"loss": 5.8783,
|
| 65 |
+
"step": 240
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.08174386920980926,
|
| 69 |
+
"grad_norm": 2.3222858905792236,
|
| 70 |
+
"learning_rate": 4.8637602179836515e-05,
|
| 71 |
+
"loss": 5.825,
|
| 72 |
+
"step": 270
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.09082652134423251,
|
| 76 |
+
"grad_norm": 2.557065010070801,
|
| 77 |
+
"learning_rate": 4.8486224644262796e-05,
|
| 78 |
+
"loss": 5.76,
|
| 79 |
+
"step": 300
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.09990917347865577,
|
| 83 |
+
"grad_norm": 2.4016597270965576,
|
| 84 |
+
"learning_rate": 4.833484710868907e-05,
|
| 85 |
+
"loss": 5.7039,
|
| 86 |
+
"step": 330
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.10899182561307902,
|
| 90 |
+
"grad_norm": 2.6895477771759033,
|
| 91 |
+
"learning_rate": 4.818346957311535e-05,
|
| 92 |
+
"loss": 5.5843,
|
| 93 |
+
"step": 360
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.11807447774750227,
|
| 97 |
+
"grad_norm": 2.741234064102173,
|
| 98 |
+
"learning_rate": 4.8032092037541634e-05,
|
| 99 |
+
"loss": 5.6376,
|
| 100 |
+
"step": 390
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1271571298819255,
|
| 104 |
+
"grad_norm": 2.8266804218292236,
|
| 105 |
+
"learning_rate": 4.788071450196791e-05,
|
| 106 |
+
"loss": 5.5649,
|
| 107 |
+
"step": 420
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.1362397820163488,
|
| 111 |
+
"grad_norm": 2.792654275894165,
|
| 112 |
+
"learning_rate": 4.772933696639419e-05,
|
| 113 |
+
"loss": 5.3651,
|
| 114 |
+
"step": 450
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.14532243415077203,
|
| 118 |
+
"grad_norm": 2.7088894844055176,
|
| 119 |
+
"learning_rate": 4.757795943082047e-05,
|
| 120 |
+
"loss": 5.4921,
|
| 121 |
+
"step": 480
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.15440508628519528,
|
| 125 |
+
"grad_norm": 2.627201795578003,
|
| 126 |
+
"learning_rate": 4.7426581895246746e-05,
|
| 127 |
+
"loss": 5.461,
|
| 128 |
+
"step": 510
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.16348773841961853,
|
| 132 |
+
"grad_norm": 2.6373610496520996,
|
| 133 |
+
"learning_rate": 4.727520435967303e-05,
|
| 134 |
+
"loss": 5.3973,
|
| 135 |
+
"step": 540
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.17257039055404177,
|
| 139 |
+
"grad_norm": 2.772226095199585,
|
| 140 |
+
"learning_rate": 4.71238268240993e-05,
|
| 141 |
+
"loss": 5.3618,
|
| 142 |
+
"step": 570
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.18165304268846502,
|
| 146 |
+
"grad_norm": 2.6005172729492188,
|
| 147 |
+
"learning_rate": 4.6972449288525583e-05,
|
| 148 |
+
"loss": 5.4365,
|
| 149 |
+
"step": 600
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.1907356948228883,
|
| 153 |
+
"grad_norm": 4.7815260887146,
|
| 154 |
+
"learning_rate": 4.6821071752951865e-05,
|
| 155 |
+
"loss": 5.3225,
|
| 156 |
+
"step": 630
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.19981834695731154,
|
| 160 |
+
"grad_norm": 2.5871763229370117,
|
| 161 |
+
"learning_rate": 4.6669694217378146e-05,
|
| 162 |
+
"loss": 5.3615,
|
| 163 |
+
"step": 660
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2089009990917348,
|
| 167 |
+
"grad_norm": 2.686840534210205,
|
| 168 |
+
"learning_rate": 4.651831668180443e-05,
|
| 169 |
+
"loss": 5.3201,
|
| 170 |
+
"step": 690
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.21798365122615804,
|
| 174 |
+
"grad_norm": 2.6963067054748535,
|
| 175 |
+
"learning_rate": 4.63669391462307e-05,
|
| 176 |
+
"loss": 5.1972,
|
| 177 |
+
"step": 720
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.22706630336058128,
|
| 181 |
+
"grad_norm": 2.9284744262695312,
|
| 182 |
+
"learning_rate": 4.6215561610656984e-05,
|
| 183 |
+
"loss": 5.3031,
|
| 184 |
+
"step": 750
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.23614895549500453,
|
| 188 |
+
"grad_norm": 2.7302122116088867,
|
| 189 |
+
"learning_rate": 4.606418407508326e-05,
|
| 190 |
+
"loss": 5.2057,
|
| 191 |
+
"step": 780
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2452316076294278,
|
| 195 |
+
"grad_norm": 2.5760107040405273,
|
| 196 |
+
"learning_rate": 4.591280653950954e-05,
|
| 197 |
+
"loss": 5.1767,
|
| 198 |
+
"step": 810
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.254314259763851,
|
| 202 |
+
"grad_norm": 2.9804234504699707,
|
| 203 |
+
"learning_rate": 4.576142900393582e-05,
|
| 204 |
+
"loss": 5.1875,
|
| 205 |
+
"step": 840
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.2633969118982743,
|
| 209 |
+
"grad_norm": 3.311448812484741,
|
| 210 |
+
"learning_rate": 4.5610051468362096e-05,
|
| 211 |
+
"loss": 5.0712,
|
| 212 |
+
"step": 870
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.2724795640326976,
|
| 216 |
+
"grad_norm": 2.67448091506958,
|
| 217 |
+
"learning_rate": 4.545867393278838e-05,
|
| 218 |
+
"loss": 5.1241,
|
| 219 |
+
"step": 900
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.2815622161671208,
|
| 223 |
+
"grad_norm": 2.8352444171905518,
|
| 224 |
+
"learning_rate": 4.530729639721465e-05,
|
| 225 |
+
"loss": 5.1732,
|
| 226 |
+
"step": 930
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.29064486830154407,
|
| 230 |
+
"grad_norm": 2.5969910621643066,
|
| 231 |
+
"learning_rate": 4.515591886164093e-05,
|
| 232 |
+
"loss": 5.0828,
|
| 233 |
+
"step": 960
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.2997275204359673,
|
| 237 |
+
"grad_norm": 2.8792121410369873,
|
| 238 |
+
"learning_rate": 4.5004541326067215e-05,
|
| 239 |
+
"loss": 5.0844,
|
| 240 |
+
"step": 990
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.30881017257039056,
|
| 244 |
+
"grad_norm": 2.9506993293762207,
|
| 245 |
+
"learning_rate": 4.485316379049349e-05,
|
| 246 |
+
"loss": 5.1764,
|
| 247 |
+
"step": 1020
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.3178928247048138,
|
| 251 |
+
"grad_norm": 2.8818390369415283,
|
| 252 |
+
"learning_rate": 4.470178625491977e-05,
|
| 253 |
+
"loss": 5.0663,
|
| 254 |
+
"step": 1050
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.32697547683923706,
|
| 258 |
+
"grad_norm": 3.128511667251587,
|
| 259 |
+
"learning_rate": 4.4550408719346046e-05,
|
| 260 |
+
"loss": 5.1026,
|
| 261 |
+
"step": 1080
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.33605812897366033,
|
| 265 |
+
"grad_norm": 3.0155856609344482,
|
| 266 |
+
"learning_rate": 4.4399031183772334e-05,
|
| 267 |
+
"loss": 5.0686,
|
| 268 |
+
"step": 1110
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.34514078110808355,
|
| 272 |
+
"grad_norm": 2.811448097229004,
|
| 273 |
+
"learning_rate": 4.424765364819861e-05,
|
| 274 |
+
"loss": 5.0351,
|
| 275 |
+
"step": 1140
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.3542234332425068,
|
| 279 |
+
"grad_norm": 2.9916000366210938,
|
| 280 |
+
"learning_rate": 4.409627611262489e-05,
|
| 281 |
+
"loss": 5.1651,
|
| 282 |
+
"step": 1170
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.36330608537693004,
|
| 286 |
+
"grad_norm": 2.9689950942993164,
|
| 287 |
+
"learning_rate": 4.394489857705117e-05,
|
| 288 |
+
"loss": 5.1457,
|
| 289 |
+
"step": 1200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.3723887375113533,
|
| 293 |
+
"grad_norm": 2.7896862030029297,
|
| 294 |
+
"learning_rate": 4.3793521041477446e-05,
|
| 295 |
+
"loss": 5.0049,
|
| 296 |
+
"step": 1230
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.3814713896457766,
|
| 300 |
+
"grad_norm": 2.790712833404541,
|
| 301 |
+
"learning_rate": 4.364214350590373e-05,
|
| 302 |
+
"loss": 4.9943,
|
| 303 |
+
"step": 1260
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.3905540417801998,
|
| 307 |
+
"grad_norm": 2.9977900981903076,
|
| 308 |
+
"learning_rate": 4.349076597033e-05,
|
| 309 |
+
"loss": 4.996,
|
| 310 |
+
"step": 1290
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.3996366939146231,
|
| 314 |
+
"grad_norm": 3.504183530807495,
|
| 315 |
+
"learning_rate": 4.333938843475628e-05,
|
| 316 |
+
"loss": 4.9611,
|
| 317 |
+
"step": 1320
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4087193460490463,
|
| 321 |
+
"grad_norm": 2.737821578979492,
|
| 322 |
+
"learning_rate": 4.3188010899182565e-05,
|
| 323 |
+
"loss": 4.9541,
|
| 324 |
+
"step": 1350
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.4178019981834696,
|
| 328 |
+
"grad_norm": 3.0585217475891113,
|
| 329 |
+
"learning_rate": 4.303663336360884e-05,
|
| 330 |
+
"loss": 4.9014,
|
| 331 |
+
"step": 1380
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.4268846503178928,
|
| 335 |
+
"grad_norm": 3.004413604736328,
|
| 336 |
+
"learning_rate": 4.288525582803512e-05,
|
| 337 |
+
"loss": 4.9703,
|
| 338 |
+
"step": 1410
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.4359673024523161,
|
| 342 |
+
"grad_norm": 2.9328274726867676,
|
| 343 |
+
"learning_rate": 4.27338782924614e-05,
|
| 344 |
+
"loss": 4.9637,
|
| 345 |
+
"step": 1440
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.44504995458673935,
|
| 349 |
+
"grad_norm": 2.93721604347229,
|
| 350 |
+
"learning_rate": 4.258250075688768e-05,
|
| 351 |
+
"loss": 4.8024,
|
| 352 |
+
"step": 1470
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.45413260672116257,
|
| 356 |
+
"grad_norm": 3.0333001613616943,
|
| 357 |
+
"learning_rate": 4.243112322131396e-05,
|
| 358 |
+
"loss": 4.8555,
|
| 359 |
+
"step": 1500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.46321525885558584,
|
| 363 |
+
"grad_norm": 3.3445775508880615,
|
| 364 |
+
"learning_rate": 4.227974568574024e-05,
|
| 365 |
+
"loss": 4.8035,
|
| 366 |
+
"step": 1530
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.47229791099000906,
|
| 370 |
+
"grad_norm": 2.9364359378814697,
|
| 371 |
+
"learning_rate": 4.212836815016652e-05,
|
| 372 |
+
"loss": 4.9296,
|
| 373 |
+
"step": 1560
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.48138056312443234,
|
| 377 |
+
"grad_norm": 2.755453586578369,
|
| 378 |
+
"learning_rate": 4.1976990614592796e-05,
|
| 379 |
+
"loss": 4.8051,
|
| 380 |
+
"step": 1590
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.4904632152588556,
|
| 384 |
+
"grad_norm": 3.0365066528320312,
|
| 385 |
+
"learning_rate": 4.182561307901908e-05,
|
| 386 |
+
"loss": 4.7833,
|
| 387 |
+
"step": 1620
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.49954586739327883,
|
| 391 |
+
"grad_norm": 3.2632575035095215,
|
| 392 |
+
"learning_rate": 4.167423554344536e-05,
|
| 393 |
+
"loss": 4.837,
|
| 394 |
+
"step": 1650
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.508628519527702,
|
| 398 |
+
"grad_norm": 3.310817003250122,
|
| 399 |
+
"learning_rate": 4.152285800787163e-05,
|
| 400 |
+
"loss": 4.7417,
|
| 401 |
+
"step": 1680
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5177111716621253,
|
| 405 |
+
"grad_norm": 3.121156692504883,
|
| 406 |
+
"learning_rate": 4.1371480472297915e-05,
|
| 407 |
+
"loss": 4.7791,
|
| 408 |
+
"step": 1710
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5267938237965486,
|
| 412 |
+
"grad_norm": 3.200591564178467,
|
| 413 |
+
"learning_rate": 4.122010293672419e-05,
|
| 414 |
+
"loss": 4.8619,
|
| 415 |
+
"step": 1740
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.5358764759309719,
|
| 419 |
+
"grad_norm": 3.1420202255249023,
|
| 420 |
+
"learning_rate": 4.106872540115047e-05,
|
| 421 |
+
"loss": 4.7576,
|
| 422 |
+
"step": 1770
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.5449591280653951,
|
| 426 |
+
"grad_norm": 3.2239160537719727,
|
| 427 |
+
"learning_rate": 4.091734786557675e-05,
|
| 428 |
+
"loss": 4.7767,
|
| 429 |
+
"step": 1800
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.5540417801998183,
|
| 433 |
+
"grad_norm": 2.9624414443969727,
|
| 434 |
+
"learning_rate": 4.076597033000303e-05,
|
| 435 |
+
"loss": 4.8608,
|
| 436 |
+
"step": 1830
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.5631244323342416,
|
| 440 |
+
"grad_norm": 3.14367938041687,
|
| 441 |
+
"learning_rate": 4.061459279442931e-05,
|
| 442 |
+
"loss": 4.7909,
|
| 443 |
+
"step": 1860
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.5722070844686649,
|
| 447 |
+
"grad_norm": 3.664564371109009,
|
| 448 |
+
"learning_rate": 4.046321525885558e-05,
|
| 449 |
+
"loss": 4.7325,
|
| 450 |
+
"step": 1890
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.5812897366030881,
|
| 454 |
+
"grad_norm": 2.9251296520233154,
|
| 455 |
+
"learning_rate": 4.0311837723281864e-05,
|
| 456 |
+
"loss": 4.8017,
|
| 457 |
+
"step": 1920
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.5903723887375113,
|
| 461 |
+
"grad_norm": 2.8796215057373047,
|
| 462 |
+
"learning_rate": 4.0160460187708146e-05,
|
| 463 |
+
"loss": 4.7124,
|
| 464 |
+
"step": 1950
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.5994550408719346,
|
| 468 |
+
"grad_norm": 3.0257513523101807,
|
| 469 |
+
"learning_rate": 4.000908265213443e-05,
|
| 470 |
+
"loss": 4.7311,
|
| 471 |
+
"step": 1980
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6085376930063578,
|
| 475 |
+
"grad_norm": 3.096799612045288,
|
| 476 |
+
"learning_rate": 3.985770511656071e-05,
|
| 477 |
+
"loss": 4.6568,
|
| 478 |
+
"step": 2010
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6176203451407811,
|
| 482 |
+
"grad_norm": 3.1430232524871826,
|
| 483 |
+
"learning_rate": 3.970632758098698e-05,
|
| 484 |
+
"loss": 4.6451,
|
| 485 |
+
"step": 2040
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.6267029972752044,
|
| 489 |
+
"grad_norm": 3.0216684341430664,
|
| 490 |
+
"learning_rate": 3.9554950045413265e-05,
|
| 491 |
+
"loss": 4.6565,
|
| 492 |
+
"step": 2070
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.6357856494096276,
|
| 496 |
+
"grad_norm": 3.0199525356292725,
|
| 497 |
+
"learning_rate": 3.940357250983954e-05,
|
| 498 |
+
"loss": 4.6988,
|
| 499 |
+
"step": 2100
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.6448683015440508,
|
| 503 |
+
"grad_norm": 2.9998953342437744,
|
| 504 |
+
"learning_rate": 3.925219497426582e-05,
|
| 505 |
+
"loss": 4.6654,
|
| 506 |
+
"step": 2130
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.6539509536784741,
|
| 510 |
+
"grad_norm": 3.15533447265625,
|
| 511 |
+
"learning_rate": 3.91008174386921e-05,
|
| 512 |
+
"loss": 4.616,
|
| 513 |
+
"step": 2160
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.6630336058128974,
|
| 517 |
+
"grad_norm": 2.8745930194854736,
|
| 518 |
+
"learning_rate": 3.894943990311838e-05,
|
| 519 |
+
"loss": 4.649,
|
| 520 |
+
"step": 2190
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.6721162579473207,
|
| 524 |
+
"grad_norm": 3.0759665966033936,
|
| 525 |
+
"learning_rate": 3.879806236754466e-05,
|
| 526 |
+
"loss": 4.6054,
|
| 527 |
+
"step": 2220
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.6811989100817438,
|
| 531 |
+
"grad_norm": 3.0508482456207275,
|
| 532 |
+
"learning_rate": 3.864668483197093e-05,
|
| 533 |
+
"loss": 4.4922,
|
| 534 |
+
"step": 2250
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.6902815622161671,
|
| 538 |
+
"grad_norm": 2.9260127544403076,
|
| 539 |
+
"learning_rate": 3.8495307296397214e-05,
|
| 540 |
+
"loss": 4.6469,
|
| 541 |
+
"step": 2280
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.6993642143505904,
|
| 545 |
+
"grad_norm": 2.924952268600464,
|
| 546 |
+
"learning_rate": 3.8343929760823496e-05,
|
| 547 |
+
"loss": 4.6164,
|
| 548 |
+
"step": 2310
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7084468664850136,
|
| 552 |
+
"grad_norm": 3.056288480758667,
|
| 553 |
+
"learning_rate": 3.819255222524977e-05,
|
| 554 |
+
"loss": 4.5877,
|
| 555 |
+
"step": 2340
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.7175295186194369,
|
| 559 |
+
"grad_norm": 4.257227420806885,
|
| 560 |
+
"learning_rate": 3.804117468967605e-05,
|
| 561 |
+
"loss": 4.6301,
|
| 562 |
+
"step": 2370
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.7266121707538601,
|
| 566 |
+
"grad_norm": 3.282137155532837,
|
| 567 |
+
"learning_rate": 3.788979715410233e-05,
|
| 568 |
+
"loss": 4.4623,
|
| 569 |
+
"step": 2400
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.7356948228882834,
|
| 573 |
+
"grad_norm": 2.945059299468994,
|
| 574 |
+
"learning_rate": 3.7738419618528615e-05,
|
| 575 |
+
"loss": 4.6267,
|
| 576 |
+
"step": 2430
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.7447774750227066,
|
| 580 |
+
"grad_norm": 3.1374645233154297,
|
| 581 |
+
"learning_rate": 3.7587042082954896e-05,
|
| 582 |
+
"loss": 4.6835,
|
| 583 |
+
"step": 2460
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.7538601271571299,
|
| 587 |
+
"grad_norm": 3.21016001701355,
|
| 588 |
+
"learning_rate": 3.743566454738117e-05,
|
| 589 |
+
"loss": 4.5581,
|
| 590 |
+
"step": 2490
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.7629427792915532,
|
| 594 |
+
"grad_norm": 2.8072383403778076,
|
| 595 |
+
"learning_rate": 3.728428701180745e-05,
|
| 596 |
+
"loss": 4.571,
|
| 597 |
+
"step": 2520
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.7720254314259763,
|
| 601 |
+
"grad_norm": 2.9735002517700195,
|
| 602 |
+
"learning_rate": 3.713290947623373e-05,
|
| 603 |
+
"loss": 4.5013,
|
| 604 |
+
"step": 2550
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.7811080835603996,
|
| 608 |
+
"grad_norm": 3.182706832885742,
|
| 609 |
+
"learning_rate": 3.698153194066001e-05,
|
| 610 |
+
"loss": 4.534,
|
| 611 |
+
"step": 2580
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.7901907356948229,
|
| 615 |
+
"grad_norm": 2.958193778991699,
|
| 616 |
+
"learning_rate": 3.683015440508629e-05,
|
| 617 |
+
"loss": 4.5697,
|
| 618 |
+
"step": 2610
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.7992733878292462,
|
| 622 |
+
"grad_norm": 2.950946569442749,
|
| 623 |
+
"learning_rate": 3.6678776869512564e-05,
|
| 624 |
+
"loss": 4.6066,
|
| 625 |
+
"step": 2640
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.8083560399636693,
|
| 629 |
+
"grad_norm": 2.9701859951019287,
|
| 630 |
+
"learning_rate": 3.6527399333938846e-05,
|
| 631 |
+
"loss": 4.5934,
|
| 632 |
+
"step": 2670
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.8174386920980926,
|
| 636 |
+
"grad_norm": 3.2177681922912598,
|
| 637 |
+
"learning_rate": 3.637602179836512e-05,
|
| 638 |
+
"loss": 4.5418,
|
| 639 |
+
"step": 2700
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.8265213442325159,
|
| 643 |
+
"grad_norm": 2.7435505390167236,
|
| 644 |
+
"learning_rate": 3.62246442627914e-05,
|
| 645 |
+
"loss": 4.5485,
|
| 646 |
+
"step": 2730
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.8356039963669392,
|
| 650 |
+
"grad_norm": 3.4409849643707275,
|
| 651 |
+
"learning_rate": 3.607326672721768e-05,
|
| 652 |
+
"loss": 4.4268,
|
| 653 |
+
"step": 2760
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.8446866485013624,
|
| 657 |
+
"grad_norm": 3.803256034851074,
|
| 658 |
+
"learning_rate": 3.592188919164396e-05,
|
| 659 |
+
"loss": 4.5643,
|
| 660 |
+
"step": 2790
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.8537693006357856,
|
| 664 |
+
"grad_norm": 3.0399341583251953,
|
| 665 |
+
"learning_rate": 3.5770511656070246e-05,
|
| 666 |
+
"loss": 4.4783,
|
| 667 |
+
"step": 2820
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.8628519527702089,
|
| 671 |
+
"grad_norm": 2.9948980808258057,
|
| 672 |
+
"learning_rate": 3.561913412049652e-05,
|
| 673 |
+
"loss": 4.4929,
|
| 674 |
+
"step": 2850
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.8719346049046321,
|
| 678 |
+
"grad_norm": 3.400299549102783,
|
| 679 |
+
"learning_rate": 3.54677565849228e-05,
|
| 680 |
+
"loss": 4.4803,
|
| 681 |
+
"step": 2880
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.8810172570390554,
|
| 685 |
+
"grad_norm": 2.9282257556915283,
|
| 686 |
+
"learning_rate": 3.531637904934908e-05,
|
| 687 |
+
"loss": 4.4554,
|
| 688 |
+
"step": 2910
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.8900999091734787,
|
| 692 |
+
"grad_norm": 2.957598924636841,
|
| 693 |
+
"learning_rate": 3.516500151377536e-05,
|
| 694 |
+
"loss": 4.5324,
|
| 695 |
+
"step": 2940
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.8991825613079019,
|
| 699 |
+
"grad_norm": 2.9992153644561768,
|
| 700 |
+
"learning_rate": 3.501362397820164e-05,
|
| 701 |
+
"loss": 4.508,
|
| 702 |
+
"step": 2970
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.9082652134423251,
|
| 706 |
+
"grad_norm": 3.1509618759155273,
|
| 707 |
+
"learning_rate": 3.4862246442627914e-05,
|
| 708 |
+
"loss": 4.4265,
|
| 709 |
+
"step": 3000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.9173478655767484,
|
| 713 |
+
"grad_norm": 3.027726888656616,
|
| 714 |
+
"learning_rate": 3.4710868907054196e-05,
|
| 715 |
+
"loss": 4.4979,
|
| 716 |
+
"step": 3030
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.9264305177111717,
|
| 720 |
+
"grad_norm": 3.0711803436279297,
|
| 721 |
+
"learning_rate": 3.455949137148047e-05,
|
| 722 |
+
"loss": 4.4946,
|
| 723 |
+
"step": 3060
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.935513169845595,
|
| 727 |
+
"grad_norm": 2.982269287109375,
|
| 728 |
+
"learning_rate": 3.440811383590675e-05,
|
| 729 |
+
"loss": 4.3433,
|
| 730 |
+
"step": 3090
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.9445958219800181,
|
| 734 |
+
"grad_norm": 2.9734480381011963,
|
| 735 |
+
"learning_rate": 3.425673630033303e-05,
|
| 736 |
+
"loss": 4.453,
|
| 737 |
+
"step": 3120
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.9536784741144414,
|
| 741 |
+
"grad_norm": 2.985030174255371,
|
| 742 |
+
"learning_rate": 3.410535876475931e-05,
|
| 743 |
+
"loss": 4.3705,
|
| 744 |
+
"step": 3150
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.9627611262488647,
|
| 748 |
+
"grad_norm": 3.1812829971313477,
|
| 749 |
+
"learning_rate": 3.395398122918559e-05,
|
| 750 |
+
"loss": 4.3414,
|
| 751 |
+
"step": 3180
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.971843778383288,
|
| 755 |
+
"grad_norm": 3.415923595428467,
|
| 756 |
+
"learning_rate": 3.380260369361187e-05,
|
| 757 |
+
"loss": 4.522,
|
| 758 |
+
"step": 3210
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.9809264305177112,
|
| 762 |
+
"grad_norm": 3.176737070083618,
|
| 763 |
+
"learning_rate": 3.3651226158038145e-05,
|
| 764 |
+
"loss": 4.4112,
|
| 765 |
+
"step": 3240
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.9900090826521344,
|
| 769 |
+
"grad_norm": 3.1306254863739014,
|
| 770 |
+
"learning_rate": 3.3499848622464433e-05,
|
| 771 |
+
"loss": 4.5104,
|
| 772 |
+
"step": 3270
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.9990917347865577,
|
| 776 |
+
"grad_norm": 3.216395616531372,
|
| 777 |
+
"learning_rate": 3.334847108689071e-05,
|
| 778 |
+
"loss": 4.3244,
|
| 779 |
+
"step": 3300
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.008174386920981,
|
| 783 |
+
"grad_norm": 3.1889307498931885,
|
| 784 |
+
"learning_rate": 3.319709355131699e-05,
|
| 785 |
+
"loss": 4.3521,
|
| 786 |
+
"step": 3330
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.017257039055404,
|
| 790 |
+
"grad_norm": 2.8001787662506104,
|
| 791 |
+
"learning_rate": 3.3045716015743264e-05,
|
| 792 |
+
"loss": 4.3047,
|
| 793 |
+
"step": 3360
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.0263396911898275,
|
| 797 |
+
"grad_norm": 3.5796685218811035,
|
| 798 |
+
"learning_rate": 3.2894338480169546e-05,
|
| 799 |
+
"loss": 4.1921,
|
| 800 |
+
"step": 3390
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.0354223433242506,
|
| 804 |
+
"grad_norm": 3.725538730621338,
|
| 805 |
+
"learning_rate": 3.274296094459583e-05,
|
| 806 |
+
"loss": 4.3203,
|
| 807 |
+
"step": 3420
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.044504995458674,
|
| 811 |
+
"grad_norm": 2.9058167934417725,
|
| 812 |
+
"learning_rate": 3.25915834090221e-05,
|
| 813 |
+
"loss": 4.385,
|
| 814 |
+
"step": 3450
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.0535876475930972,
|
| 818 |
+
"grad_norm": 3.120119333267212,
|
| 819 |
+
"learning_rate": 3.244020587344838e-05,
|
| 820 |
+
"loss": 4.2883,
|
| 821 |
+
"step": 3480
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.0626702997275204,
|
| 825 |
+
"grad_norm": 3.230036735534668,
|
| 826 |
+
"learning_rate": 3.228882833787466e-05,
|
| 827 |
+
"loss": 4.3602,
|
| 828 |
+
"step": 3510
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.0717529518619437,
|
| 832 |
+
"grad_norm": 3.482921600341797,
|
| 833 |
+
"learning_rate": 3.213745080230094e-05,
|
| 834 |
+
"loss": 4.3984,
|
| 835 |
+
"step": 3540
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.080835603996367,
|
| 839 |
+
"grad_norm": 3.0121572017669678,
|
| 840 |
+
"learning_rate": 3.198607326672722e-05,
|
| 841 |
+
"loss": 4.3864,
|
| 842 |
+
"step": 3570
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.0899182561307903,
|
| 846 |
+
"grad_norm": 3.277411460876465,
|
| 847 |
+
"learning_rate": 3.1834695731153495e-05,
|
| 848 |
+
"loss": 4.2294,
|
| 849 |
+
"step": 3600
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.0990009082652135,
|
| 853 |
+
"grad_norm": 3.0383167266845703,
|
| 854 |
+
"learning_rate": 3.168331819557978e-05,
|
| 855 |
+
"loss": 4.2759,
|
| 856 |
+
"step": 3630
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.1080835603996366,
|
| 860 |
+
"grad_norm": 3.3026745319366455,
|
| 861 |
+
"learning_rate": 3.153194066000605e-05,
|
| 862 |
+
"loss": 4.3093,
|
| 863 |
+
"step": 3660
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.11716621253406,
|
| 867 |
+
"grad_norm": 2.954747200012207,
|
| 868 |
+
"learning_rate": 3.138056312443234e-05,
|
| 869 |
+
"loss": 4.2476,
|
| 870 |
+
"step": 3690
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.1262488646684832,
|
| 874 |
+
"grad_norm": 3.2137765884399414,
|
| 875 |
+
"learning_rate": 3.1229185588858614e-05,
|
| 876 |
+
"loss": 4.2858,
|
| 877 |
+
"step": 3720
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.1353315168029066,
|
| 881 |
+
"grad_norm": 3.4028799533843994,
|
| 882 |
+
"learning_rate": 3.1077808053284896e-05,
|
| 883 |
+
"loss": 4.3652,
|
| 884 |
+
"step": 3750
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.1444141689373297,
|
| 888 |
+
"grad_norm": 3.0039563179016113,
|
| 889 |
+
"learning_rate": 3.092643051771118e-05,
|
| 890 |
+
"loss": 4.4106,
|
| 891 |
+
"step": 3780
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.1534968210717529,
|
| 895 |
+
"grad_norm": 2.973820209503174,
|
| 896 |
+
"learning_rate": 3.077505298213745e-05,
|
| 897 |
+
"loss": 4.1827,
|
| 898 |
+
"step": 3810
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.1625794732061763,
|
| 902 |
+
"grad_norm": 2.99037766456604,
|
| 903 |
+
"learning_rate": 3.062367544656373e-05,
|
| 904 |
+
"loss": 4.3092,
|
| 905 |
+
"step": 3840
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.1716621253405994,
|
| 909 |
+
"grad_norm": 3.181398391723633,
|
| 910 |
+
"learning_rate": 3.047229791099001e-05,
|
| 911 |
+
"loss": 4.417,
|
| 912 |
+
"step": 3870
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.1807447774750228,
|
| 916 |
+
"grad_norm": 3.1933484077453613,
|
| 917 |
+
"learning_rate": 3.032092037541629e-05,
|
| 918 |
+
"loss": 4.2361,
|
| 919 |
+
"step": 3900
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.189827429609446,
|
| 923 |
+
"grad_norm": 3.4427855014801025,
|
| 924 |
+
"learning_rate": 3.0169542839842567e-05,
|
| 925 |
+
"loss": 4.2687,
|
| 926 |
+
"step": 3930
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.1989100817438691,
|
| 930 |
+
"grad_norm": 3.0683298110961914,
|
| 931 |
+
"learning_rate": 3.001816530426885e-05,
|
| 932 |
+
"loss": 4.2748,
|
| 933 |
+
"step": 3960
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.2079927338782925,
|
| 937 |
+
"grad_norm": 3.044698715209961,
|
| 938 |
+
"learning_rate": 2.9866787768695127e-05,
|
| 939 |
+
"loss": 4.2671,
|
| 940 |
+
"step": 3990
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.2170753860127157,
|
| 944 |
+
"grad_norm": 3.1354904174804688,
|
| 945 |
+
"learning_rate": 2.9715410233121405e-05,
|
| 946 |
+
"loss": 4.2635,
|
| 947 |
+
"step": 4020
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.226158038147139,
|
| 951 |
+
"grad_norm": 3.282745361328125,
|
| 952 |
+
"learning_rate": 2.9564032697547683e-05,
|
| 953 |
+
"loss": 4.3544,
|
| 954 |
+
"step": 4050
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.2352406902815622,
|
| 958 |
+
"grad_norm": 3.369798183441162,
|
| 959 |
+
"learning_rate": 2.941265516197396e-05,
|
| 960 |
+
"loss": 4.1993,
|
| 961 |
+
"step": 4080
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.2443233424159854,
|
| 965 |
+
"grad_norm": 3.395785331726074,
|
| 966 |
+
"learning_rate": 2.9261277626400242e-05,
|
| 967 |
+
"loss": 4.1131,
|
| 968 |
+
"step": 4110
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.2534059945504088,
|
| 972 |
+
"grad_norm": 3.500697135925293,
|
| 973 |
+
"learning_rate": 2.9109900090826524e-05,
|
| 974 |
+
"loss": 4.192,
|
| 975 |
+
"step": 4140
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.262488646684832,
|
| 979 |
+
"grad_norm": 2.94278621673584,
|
| 980 |
+
"learning_rate": 2.8958522555252805e-05,
|
| 981 |
+
"loss": 4.2863,
|
| 982 |
+
"step": 4170
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.2715712988192553,
|
| 986 |
+
"grad_norm": 3.3217315673828125,
|
| 987 |
+
"learning_rate": 2.8807145019679083e-05,
|
| 988 |
+
"loss": 4.1763,
|
| 989 |
+
"step": 4200
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.2806539509536785,
|
| 993 |
+
"grad_norm": 3.232830762863159,
|
| 994 |
+
"learning_rate": 2.865576748410536e-05,
|
| 995 |
+
"loss": 4.2595,
|
| 996 |
+
"step": 4230
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.2897366030881017,
|
| 1000 |
+
"grad_norm": 3.3042378425598145,
|
| 1001 |
+
"learning_rate": 2.850438994853164e-05,
|
| 1002 |
+
"loss": 4.2393,
|
| 1003 |
+
"step": 4260
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.298819255222525,
|
| 1007 |
+
"grad_norm": 3.83151912689209,
|
| 1008 |
+
"learning_rate": 2.835301241295792e-05,
|
| 1009 |
+
"loss": 4.3005,
|
| 1010 |
+
"step": 4290
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.3079019073569482,
|
| 1014 |
+
"grad_norm": 3.245086431503296,
|
| 1015 |
+
"learning_rate": 2.82016348773842e-05,
|
| 1016 |
+
"loss": 4.205,
|
| 1017 |
+
"step": 4320
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.3169845594913716,
|
| 1021 |
+
"grad_norm": 3.4392285346984863,
|
| 1022 |
+
"learning_rate": 2.8050257341810477e-05,
|
| 1023 |
+
"loss": 4.1964,
|
| 1024 |
+
"step": 4350
|
| 1025 |
+
}
|
| 1026 |
+
],
|
| 1027 |
+
"logging_steps": 30,
|
| 1028 |
+
"max_steps": 9909,
|
| 1029 |
+
"num_input_tokens_seen": 0,
|
| 1030 |
+
"num_train_epochs": 3,
|
| 1031 |
+
"save_steps": 30,
|
| 1032 |
+
"total_flos": 1136555016192000.0,
|
| 1033 |
+
"train_batch_size": 4,
|
| 1034 |
+
"trial_name": null,
|
| 1035 |
+
"trial_params": null
|
| 1036 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51073e9acaf31db2dd04a38614739acc9b8272913419c8770c1a7b85bb4facb3
|
| 3 |
+
size 4920
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|