iko v5 here!!
Browse files- config.json +38 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +23 -0
- trainer_state.json +2352 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_ctx": 1024,
|
| 14 |
+
"n_embd": 768,
|
| 15 |
+
"n_head": 12,
|
| 16 |
+
"n_inner": null,
|
| 17 |
+
"n_layer": 12,
|
| 18 |
+
"n_positions": 1024,
|
| 19 |
+
"reorder_and_upcast_attn": false,
|
| 20 |
+
"resid_pdrop": 0.1,
|
| 21 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 22 |
+
"scale_attn_weights": true,
|
| 23 |
+
"summary_activation": null,
|
| 24 |
+
"summary_first_dropout": 0.1,
|
| 25 |
+
"summary_proj_to_labels": true,
|
| 26 |
+
"summary_type": "cls_index",
|
| 27 |
+
"summary_use_proj": true,
|
| 28 |
+
"task_specific_params": {
|
| 29 |
+
"text-generation": {
|
| 30 |
+
"do_sample": true,
|
| 31 |
+
"max_length": 50
|
| 32 |
+
}
|
| 33 |
+
},
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.55.0",
|
| 36 |
+
"use_cache": true,
|
| 37 |
+
"vocab_size": 50257
|
| 38 |
+
}
|
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": "4.55.0"
|
| 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:bd370c65092ea6bea2225a7f4bf00074e37d5288d0fc5b9ef855c6060e428d3c
|
| 3 |
+
size 497774208
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ce816922be966b8b5ffccfb76823f5604f366de145d61b3b02bedc837a1e16a
|
| 3 |
+
size 995642298
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d2ea38b34dcfe0437592d06ea2ff371a7b2b24b0020868f8752b05877aa72ee
|
| 3 |
+
size 14244
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0399dea0e3276ef1798dc9583e0bf6177a9e1467fe054f3571bd76b577ebfd1b
|
| 3 |
+
size 988
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4845f59badd9a7bf95fe94a60b1515a10aad2c2de4b64472f0dfb8dd7947388b
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"50256": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
}
|
| 13 |
+
},
|
| 14 |
+
"bos_token": "<|endoftext|>",
|
| 15 |
+
"clean_up_tokenization_spaces": false,
|
| 16 |
+
"eos_token": "<|endoftext|>",
|
| 17 |
+
"errors": "replace",
|
| 18 |
+
"extra_special_tokens": {},
|
| 19 |
+
"model_max_length": 1024,
|
| 20 |
+
"pad_token": "<|endoftext|>",
|
| 21 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 22 |
+
"unk_token": "<|endoftext|>"
|
| 23 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2352 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 16503,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0030298439630359036,
|
| 14 |
+
"grad_norm": 1.5333168506622314,
|
| 15 |
+
"learning_rate": 9.970308428770528e-06,
|
| 16 |
+
"loss": 2.3086,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.006059687926071807,
|
| 21 |
+
"grad_norm": 1.427223563194275,
|
| 22 |
+
"learning_rate": 9.9400109071078e-06,
|
| 23 |
+
"loss": 2.222,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00908953188910771,
|
| 28 |
+
"grad_norm": 1.435951590538025,
|
| 29 |
+
"learning_rate": 9.909713385445072e-06,
|
| 30 |
+
"loss": 2.1361,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.012119375852143614,
|
| 35 |
+
"grad_norm": 1.7074558734893799,
|
| 36 |
+
"learning_rate": 9.879415863782343e-06,
|
| 37 |
+
"loss": 2.0743,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.015149219815179518,
|
| 42 |
+
"grad_norm": 1.5592257976531982,
|
| 43 |
+
"learning_rate": 9.849118342119616e-06,
|
| 44 |
+
"loss": 2.0526,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.01817906377821542,
|
| 49 |
+
"grad_norm": 1.5479929447174072,
|
| 50 |
+
"learning_rate": 9.818820820456887e-06,
|
| 51 |
+
"loss": 2.0152,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.021208907741251327,
|
| 56 |
+
"grad_norm": 1.8903971910476685,
|
| 57 |
+
"learning_rate": 9.78852329879416e-06,
|
| 58 |
+
"loss": 1.9692,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.02423875170428723,
|
| 63 |
+
"grad_norm": 1.7212985754013062,
|
| 64 |
+
"learning_rate": 9.758225777131432e-06,
|
| 65 |
+
"loss": 1.9869,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.027268595667323134,
|
| 70 |
+
"grad_norm": 1.5003551244735718,
|
| 71 |
+
"learning_rate": 9.727928255468703e-06,
|
| 72 |
+
"loss": 1.9526,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.030298439630359036,
|
| 77 |
+
"grad_norm": 1.938314437866211,
|
| 78 |
+
"learning_rate": 9.697630733805974e-06,
|
| 79 |
+
"loss": 1.9265,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.03332828359339494,
|
| 84 |
+
"grad_norm": 2.260500431060791,
|
| 85 |
+
"learning_rate": 9.667333212143247e-06,
|
| 86 |
+
"loss": 1.9519,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.03635812755643084,
|
| 91 |
+
"grad_norm": 1.8720899820327759,
|
| 92 |
+
"learning_rate": 9.63703569048052e-06,
|
| 93 |
+
"loss": 1.9269,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.03938797151946675,
|
| 98 |
+
"grad_norm": 1.6958003044128418,
|
| 99 |
+
"learning_rate": 9.606738168817791e-06,
|
| 100 |
+
"loss": 1.8689,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.042417815482502653,
|
| 105 |
+
"grad_norm": 1.8129528760910034,
|
| 106 |
+
"learning_rate": 9.576440647155064e-06,
|
| 107 |
+
"loss": 1.9042,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.04544765944553855,
|
| 112 |
+
"grad_norm": 1.8754149675369263,
|
| 113 |
+
"learning_rate": 9.546143125492335e-06,
|
| 114 |
+
"loss": 1.8763,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.04847750340857446,
|
| 119 |
+
"grad_norm": 1.8043217658996582,
|
| 120 |
+
"learning_rate": 9.515845603829608e-06,
|
| 121 |
+
"loss": 1.8885,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.05150734737161036,
|
| 126 |
+
"grad_norm": 1.7908295392990112,
|
| 127 |
+
"learning_rate": 9.48554808216688e-06,
|
| 128 |
+
"loss": 1.894,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.05453719133464627,
|
| 133 |
+
"grad_norm": 1.8442623615264893,
|
| 134 |
+
"learning_rate": 9.455250560504152e-06,
|
| 135 |
+
"loss": 1.8439,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.057567035297682166,
|
| 140 |
+
"grad_norm": 2.4318511486053467,
|
| 141 |
+
"learning_rate": 9.424953038841423e-06,
|
| 142 |
+
"loss": 1.8462,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.06059687926071807,
|
| 147 |
+
"grad_norm": 1.9251976013183594,
|
| 148 |
+
"learning_rate": 9.394655517178695e-06,
|
| 149 |
+
"loss": 1.8359,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.06362672322375397,
|
| 154 |
+
"grad_norm": 2.037912368774414,
|
| 155 |
+
"learning_rate": 9.364357995515968e-06,
|
| 156 |
+
"loss": 1.8454,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.06665656718678988,
|
| 161 |
+
"grad_norm": 1.7860537767410278,
|
| 162 |
+
"learning_rate": 9.33406047385324e-06,
|
| 163 |
+
"loss": 1.8139,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.06968641114982578,
|
| 168 |
+
"grad_norm": 2.082537889480591,
|
| 169 |
+
"learning_rate": 9.303762952190512e-06,
|
| 170 |
+
"loss": 1.8248,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.07271625511286169,
|
| 175 |
+
"grad_norm": 1.863715648651123,
|
| 176 |
+
"learning_rate": 9.273465430527783e-06,
|
| 177 |
+
"loss": 1.8146,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.07574609907589759,
|
| 182 |
+
"grad_norm": 1.9475610256195068,
|
| 183 |
+
"learning_rate": 9.243167908865056e-06,
|
| 184 |
+
"loss": 1.8335,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0787759430389335,
|
| 189 |
+
"grad_norm": 1.8209477663040161,
|
| 190 |
+
"learning_rate": 9.212870387202327e-06,
|
| 191 |
+
"loss": 1.841,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.0818057870019694,
|
| 196 |
+
"grad_norm": 1.7817108631134033,
|
| 197 |
+
"learning_rate": 9.1825728655396e-06,
|
| 198 |
+
"loss": 1.8559,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.08483563096500531,
|
| 203 |
+
"grad_norm": 2.1402652263641357,
|
| 204 |
+
"learning_rate": 9.152275343876873e-06,
|
| 205 |
+
"loss": 1.7597,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.08786547492804121,
|
| 210 |
+
"grad_norm": 1.7914282083511353,
|
| 211 |
+
"learning_rate": 9.121977822214144e-06,
|
| 212 |
+
"loss": 1.7555,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0908953188910771,
|
| 217 |
+
"grad_norm": 1.8525111675262451,
|
| 218 |
+
"learning_rate": 9.091680300551415e-06,
|
| 219 |
+
"loss": 1.7685,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.09392516285411301,
|
| 224 |
+
"grad_norm": 1.7323678731918335,
|
| 225 |
+
"learning_rate": 9.061382778888688e-06,
|
| 226 |
+
"loss": 1.7729,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.09695500681714891,
|
| 231 |
+
"grad_norm": 1.8842493295669556,
|
| 232 |
+
"learning_rate": 9.03108525722596e-06,
|
| 233 |
+
"loss": 1.8152,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.09998485078018482,
|
| 238 |
+
"grad_norm": 2.3138351440429688,
|
| 239 |
+
"learning_rate": 9.000787735563232e-06,
|
| 240 |
+
"loss": 1.7888,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.10301469474322072,
|
| 245 |
+
"grad_norm": 1.5812338590621948,
|
| 246 |
+
"learning_rate": 8.970490213900503e-06,
|
| 247 |
+
"loss": 1.736,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.10604453870625663,
|
| 252 |
+
"grad_norm": 1.7662309408187866,
|
| 253 |
+
"learning_rate": 8.940192692237775e-06,
|
| 254 |
+
"loss": 1.7532,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.10907438266929254,
|
| 259 |
+
"grad_norm": 1.6153583526611328,
|
| 260 |
+
"learning_rate": 8.909895170575048e-06,
|
| 261 |
+
"loss": 1.7995,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.11210422663232844,
|
| 266 |
+
"grad_norm": 1.7088361978530884,
|
| 267 |
+
"learning_rate": 8.87959764891232e-06,
|
| 268 |
+
"loss": 1.7612,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.11513407059536433,
|
| 273 |
+
"grad_norm": 1.7324771881103516,
|
| 274 |
+
"learning_rate": 8.849300127249592e-06,
|
| 275 |
+
"loss": 1.7803,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.11816391455840024,
|
| 280 |
+
"grad_norm": 2.0461599826812744,
|
| 281 |
+
"learning_rate": 8.819002605586863e-06,
|
| 282 |
+
"loss": 1.7642,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.12119375852143614,
|
| 287 |
+
"grad_norm": 1.9488065242767334,
|
| 288 |
+
"learning_rate": 8.788705083924136e-06,
|
| 289 |
+
"loss": 1.7281,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.12422360248447205,
|
| 294 |
+
"grad_norm": 1.6670262813568115,
|
| 295 |
+
"learning_rate": 8.758407562261407e-06,
|
| 296 |
+
"loss": 1.7446,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.12725344644750794,
|
| 301 |
+
"grad_norm": 2.1428778171539307,
|
| 302 |
+
"learning_rate": 8.72811004059868e-06,
|
| 303 |
+
"loss": 1.7129,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.13028329041054384,
|
| 308 |
+
"grad_norm": 1.863022804260254,
|
| 309 |
+
"learning_rate": 8.697812518935953e-06,
|
| 310 |
+
"loss": 1.7471,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.13331313437357975,
|
| 315 |
+
"grad_norm": 1.9406341314315796,
|
| 316 |
+
"learning_rate": 8.667514997273224e-06,
|
| 317 |
+
"loss": 1.7381,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.13634297833661566,
|
| 322 |
+
"grad_norm": 2.044616222381592,
|
| 323 |
+
"learning_rate": 8.637217475610495e-06,
|
| 324 |
+
"loss": 1.6925,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.13937282229965156,
|
| 329 |
+
"grad_norm": 2.1815521717071533,
|
| 330 |
+
"learning_rate": 8.606919953947768e-06,
|
| 331 |
+
"loss": 1.7528,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.14240266626268747,
|
| 336 |
+
"grad_norm": 1.943009853363037,
|
| 337 |
+
"learning_rate": 8.57662243228504e-06,
|
| 338 |
+
"loss": 1.7448,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.14543251022572337,
|
| 343 |
+
"grad_norm": 1.806156873703003,
|
| 344 |
+
"learning_rate": 8.546324910622311e-06,
|
| 345 |
+
"loss": 1.7086,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.14846235418875928,
|
| 350 |
+
"grad_norm": 1.6711620092391968,
|
| 351 |
+
"learning_rate": 8.516027388959584e-06,
|
| 352 |
+
"loss": 1.7235,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.15149219815179518,
|
| 357 |
+
"grad_norm": 1.807391881942749,
|
| 358 |
+
"learning_rate": 8.48633581773011e-06,
|
| 359 |
+
"loss": 1.7185,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.1545220421148311,
|
| 364 |
+
"grad_norm": 1.809409499168396,
|
| 365 |
+
"learning_rate": 8.456038296067382e-06,
|
| 366 |
+
"loss": 1.7292,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.157551886077867,
|
| 371 |
+
"grad_norm": 1.953994870185852,
|
| 372 |
+
"learning_rate": 8.425740774404655e-06,
|
| 373 |
+
"loss": 1.7225,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.1605817300409029,
|
| 378 |
+
"grad_norm": 1.776473045349121,
|
| 379 |
+
"learning_rate": 8.395443252741926e-06,
|
| 380 |
+
"loss": 1.707,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.1636115740039388,
|
| 385 |
+
"grad_norm": 2.0671963691711426,
|
| 386 |
+
"learning_rate": 8.365145731079199e-06,
|
| 387 |
+
"loss": 1.6829,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.1666414179669747,
|
| 392 |
+
"grad_norm": 1.7150015830993652,
|
| 393 |
+
"learning_rate": 8.33484820941647e-06,
|
| 394 |
+
"loss": 1.6755,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.16967126193001061,
|
| 399 |
+
"grad_norm": 1.7115235328674316,
|
| 400 |
+
"learning_rate": 8.304550687753743e-06,
|
| 401 |
+
"loss": 1.7345,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.17270110589304652,
|
| 406 |
+
"grad_norm": 1.6895759105682373,
|
| 407 |
+
"learning_rate": 8.274253166091014e-06,
|
| 408 |
+
"loss": 1.6911,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.17573094985608242,
|
| 413 |
+
"grad_norm": 1.8337711095809937,
|
| 414 |
+
"learning_rate": 8.243955644428287e-06,
|
| 415 |
+
"loss": 1.717,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.1787607938191183,
|
| 420 |
+
"grad_norm": 1.6573454141616821,
|
| 421 |
+
"learning_rate": 8.21365812276556e-06,
|
| 422 |
+
"loss": 1.6788,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.1817906377821542,
|
| 427 |
+
"grad_norm": 1.6749043464660645,
|
| 428 |
+
"learning_rate": 8.18336060110283e-06,
|
| 429 |
+
"loss": 1.6658,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.1848204817451901,
|
| 434 |
+
"grad_norm": 1.8717437982559204,
|
| 435 |
+
"learning_rate": 8.153063079440102e-06,
|
| 436 |
+
"loss": 1.6504,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.18785032570822602,
|
| 441 |
+
"grad_norm": 1.8825408220291138,
|
| 442 |
+
"learning_rate": 8.122765557777375e-06,
|
| 443 |
+
"loss": 1.689,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.19088016967126192,
|
| 448 |
+
"grad_norm": 1.528969645500183,
|
| 449 |
+
"learning_rate": 8.092468036114647e-06,
|
| 450 |
+
"loss": 1.6896,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.19391001363429783,
|
| 455 |
+
"grad_norm": 1.6371997594833374,
|
| 456 |
+
"learning_rate": 8.062170514451918e-06,
|
| 457 |
+
"loss": 1.6924,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.19693985759733373,
|
| 462 |
+
"grad_norm": 1.6460844278335571,
|
| 463 |
+
"learning_rate": 8.03187299278919e-06,
|
| 464 |
+
"loss": 1.7174,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.19996970156036964,
|
| 469 |
+
"grad_norm": 1.8048419952392578,
|
| 470 |
+
"learning_rate": 8.001575471126462e-06,
|
| 471 |
+
"loss": 1.6752,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.20299954552340554,
|
| 476 |
+
"grad_norm": 1.5663115978240967,
|
| 477 |
+
"learning_rate": 7.971277949463735e-06,
|
| 478 |
+
"loss": 1.67,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.20602938948644145,
|
| 483 |
+
"grad_norm": 1.651563048362732,
|
| 484 |
+
"learning_rate": 7.940980427801006e-06,
|
| 485 |
+
"loss": 1.676,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.20905923344947736,
|
| 490 |
+
"grad_norm": 1.7851747274398804,
|
| 491 |
+
"learning_rate": 7.910682906138279e-06,
|
| 492 |
+
"loss": 1.6498,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.21208907741251326,
|
| 497 |
+
"grad_norm": 1.839625358581543,
|
| 498 |
+
"learning_rate": 7.88038538447555e-06,
|
| 499 |
+
"loss": 1.6772,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.21511892137554917,
|
| 504 |
+
"grad_norm": 1.769681453704834,
|
| 505 |
+
"learning_rate": 7.850087862812823e-06,
|
| 506 |
+
"loss": 1.6827,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.21814876533858507,
|
| 511 |
+
"grad_norm": 1.9082216024398804,
|
| 512 |
+
"learning_rate": 7.819790341150094e-06,
|
| 513 |
+
"loss": 1.6918,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.22117860930162098,
|
| 518 |
+
"grad_norm": 1.7255308628082275,
|
| 519 |
+
"learning_rate": 7.789492819487367e-06,
|
| 520 |
+
"loss": 1.6702,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.22420845326465688,
|
| 525 |
+
"grad_norm": 1.5342808961868286,
|
| 526 |
+
"learning_rate": 7.75919529782464e-06,
|
| 527 |
+
"loss": 1.6514,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.2272382972276928,
|
| 532 |
+
"grad_norm": 1.6255989074707031,
|
| 533 |
+
"learning_rate": 7.72889777616191e-06,
|
| 534 |
+
"loss": 1.6619,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.23026814119072866,
|
| 539 |
+
"grad_norm": 1.726608157157898,
|
| 540 |
+
"learning_rate": 7.698600254499182e-06,
|
| 541 |
+
"loss": 1.6691,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.23329798515376457,
|
| 546 |
+
"grad_norm": 1.7505511045455933,
|
| 547 |
+
"learning_rate": 7.668302732836454e-06,
|
| 548 |
+
"loss": 1.6603,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.23632782911680048,
|
| 553 |
+
"grad_norm": 1.7649147510528564,
|
| 554 |
+
"learning_rate": 7.638005211173727e-06,
|
| 555 |
+
"loss": 1.7008,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.23935767307983638,
|
| 560 |
+
"grad_norm": 1.7847344875335693,
|
| 561 |
+
"learning_rate": 7.607707689510999e-06,
|
| 562 |
+
"loss": 1.6701,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.24238751704287229,
|
| 567 |
+
"grad_norm": 1.7049511671066284,
|
| 568 |
+
"learning_rate": 7.57741016784827e-06,
|
| 569 |
+
"loss": 1.6446,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.2454173610059082,
|
| 574 |
+
"grad_norm": 1.5229992866516113,
|
| 575 |
+
"learning_rate": 7.547112646185543e-06,
|
| 576 |
+
"loss": 1.6482,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.2484472049689441,
|
| 581 |
+
"grad_norm": 1.7003657817840576,
|
| 582 |
+
"learning_rate": 7.516815124522815e-06,
|
| 583 |
+
"loss": 1.6154,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.25147704893198003,
|
| 588 |
+
"grad_norm": 2.140820026397705,
|
| 589 |
+
"learning_rate": 7.486517602860086e-06,
|
| 590 |
+
"loss": 1.6522,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.2545068928950159,
|
| 595 |
+
"grad_norm": 1.580733060836792,
|
| 596 |
+
"learning_rate": 7.456220081197358e-06,
|
| 597 |
+
"loss": 1.6252,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.2575367368580518,
|
| 602 |
+
"grad_norm": 1.912180781364441,
|
| 603 |
+
"learning_rate": 7.425922559534631e-06,
|
| 604 |
+
"loss": 1.6581,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.2605665808210877,
|
| 609 |
+
"grad_norm": 1.6341553926467896,
|
| 610 |
+
"learning_rate": 7.395625037871903e-06,
|
| 611 |
+
"loss": 1.6569,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.2635964247841236,
|
| 616 |
+
"grad_norm": 1.6689584255218506,
|
| 617 |
+
"learning_rate": 7.365327516209174e-06,
|
| 618 |
+
"loss": 1.67,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.2666262687471595,
|
| 623 |
+
"grad_norm": 1.5784380435943604,
|
| 624 |
+
"learning_rate": 7.335029994546447e-06,
|
| 625 |
+
"loss": 1.6469,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.2696561127101954,
|
| 630 |
+
"grad_norm": 1.8624850511550903,
|
| 631 |
+
"learning_rate": 7.3047324728837186e-06,
|
| 632 |
+
"loss": 1.6654,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.2726859566732313,
|
| 637 |
+
"grad_norm": 1.8794349431991577,
|
| 638 |
+
"learning_rate": 7.274434951220991e-06,
|
| 639 |
+
"loss": 1.6662,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.2757158006362672,
|
| 644 |
+
"grad_norm": 1.8854588270187378,
|
| 645 |
+
"learning_rate": 7.2441374295582624e-06,
|
| 646 |
+
"loss": 1.6837,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.2787456445993031,
|
| 651 |
+
"grad_norm": 2.1526780128479004,
|
| 652 |
+
"learning_rate": 7.213839907895534e-06,
|
| 653 |
+
"loss": 1.6632,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.281775488562339,
|
| 658 |
+
"grad_norm": 1.9971776008605957,
|
| 659 |
+
"learning_rate": 7.183542386232807e-06,
|
| 660 |
+
"loss": 1.6099,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.28480533252537493,
|
| 665 |
+
"grad_norm": 1.598711609840393,
|
| 666 |
+
"learning_rate": 7.153244864570079e-06,
|
| 667 |
+
"loss": 1.6607,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.28783517648841084,
|
| 672 |
+
"grad_norm": 1.4923897981643677,
|
| 673 |
+
"learning_rate": 7.12294734290735e-06,
|
| 674 |
+
"loss": 1.6116,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.29086502045144674,
|
| 679 |
+
"grad_norm": 1.8050360679626465,
|
| 680 |
+
"learning_rate": 7.092649821244623e-06,
|
| 681 |
+
"loss": 1.6655,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.29389486441448265,
|
| 686 |
+
"grad_norm": 1.7371289730072021,
|
| 687 |
+
"learning_rate": 7.062352299581895e-06,
|
| 688 |
+
"loss": 1.6276,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.29692470837751855,
|
| 693 |
+
"grad_norm": 1.7267305850982666,
|
| 694 |
+
"learning_rate": 7.032054777919167e-06,
|
| 695 |
+
"loss": 1.6399,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.29995455234055446,
|
| 700 |
+
"grad_norm": 1.5485690832138062,
|
| 701 |
+
"learning_rate": 7.001757256256439e-06,
|
| 702 |
+
"loss": 1.6188,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.30298439630359036,
|
| 707 |
+
"grad_norm": 1.78826904296875,
|
| 708 |
+
"learning_rate": 6.971459734593711e-06,
|
| 709 |
+
"loss": 1.6426,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.30601424026662627,
|
| 714 |
+
"grad_norm": 1.8727006912231445,
|
| 715 |
+
"learning_rate": 6.941768163364238e-06,
|
| 716 |
+
"loss": 1.6329,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.3090440842296622,
|
| 721 |
+
"grad_norm": 1.8767826557159424,
|
| 722 |
+
"learning_rate": 6.91147064170151e-06,
|
| 723 |
+
"loss": 1.6724,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.3120739281926981,
|
| 728 |
+
"grad_norm": 1.4456415176391602,
|
| 729 |
+
"learning_rate": 6.881173120038781e-06,
|
| 730 |
+
"loss": 1.6698,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.315103772155734,
|
| 735 |
+
"grad_norm": 1.652795433998108,
|
| 736 |
+
"learning_rate": 6.8508755983760535e-06,
|
| 737 |
+
"loss": 1.6425,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.3181336161187699,
|
| 742 |
+
"grad_norm": 1.609587550163269,
|
| 743 |
+
"learning_rate": 6.8205780767133255e-06,
|
| 744 |
+
"loss": 1.5815,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.3211634600818058,
|
| 749 |
+
"grad_norm": 1.6337134838104248,
|
| 750 |
+
"learning_rate": 6.790280555050598e-06,
|
| 751 |
+
"loss": 1.6195,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.3241933040448417,
|
| 756 |
+
"grad_norm": 1.3837682008743286,
|
| 757 |
+
"learning_rate": 6.759983033387869e-06,
|
| 758 |
+
"loss": 1.6397,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.3272231480078776,
|
| 763 |
+
"grad_norm": 1.5774863958358765,
|
| 764 |
+
"learning_rate": 6.729685511725141e-06,
|
| 765 |
+
"loss": 1.6391,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.3302529919709135,
|
| 770 |
+
"grad_norm": 1.6407530307769775,
|
| 771 |
+
"learning_rate": 6.699387990062414e-06,
|
| 772 |
+
"loss": 1.6216,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.3332828359339494,
|
| 777 |
+
"grad_norm": 1.7346445322036743,
|
| 778 |
+
"learning_rate": 6.669090468399686e-06,
|
| 779 |
+
"loss": 1.593,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.3363126798969853,
|
| 784 |
+
"grad_norm": 1.907456636428833,
|
| 785 |
+
"learning_rate": 6.638792946736957e-06,
|
| 786 |
+
"loss": 1.5871,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.33934252386002123,
|
| 791 |
+
"grad_norm": 1.8702911138534546,
|
| 792 |
+
"learning_rate": 6.60849542507423e-06,
|
| 793 |
+
"loss": 1.6464,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.34237236782305713,
|
| 798 |
+
"grad_norm": 1.750834345817566,
|
| 799 |
+
"learning_rate": 6.578197903411502e-06,
|
| 800 |
+
"loss": 1.5719,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.34540221178609304,
|
| 805 |
+
"grad_norm": 1.632150650024414,
|
| 806 |
+
"learning_rate": 6.547900381748773e-06,
|
| 807 |
+
"loss": 1.6209,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.34843205574912894,
|
| 812 |
+
"grad_norm": 1.6225661039352417,
|
| 813 |
+
"learning_rate": 6.517602860086045e-06,
|
| 814 |
+
"loss": 1.6307,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.35146189971216485,
|
| 819 |
+
"grad_norm": 1.6222798824310303,
|
| 820 |
+
"learning_rate": 6.487305338423318e-06,
|
| 821 |
+
"loss": 1.6103,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.35449174367520075,
|
| 826 |
+
"grad_norm": 1.9228448867797852,
|
| 827 |
+
"learning_rate": 6.45700781676059e-06,
|
| 828 |
+
"loss": 1.6603,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.3575215876382366,
|
| 833 |
+
"grad_norm": 1.9268038272857666,
|
| 834 |
+
"learning_rate": 6.426710295097861e-06,
|
| 835 |
+
"loss": 1.6057,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.3605514316012725,
|
| 840 |
+
"grad_norm": 1.4592260122299194,
|
| 841 |
+
"learning_rate": 6.3964127734351335e-06,
|
| 842 |
+
"loss": 1.639,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.3635812755643084,
|
| 847 |
+
"grad_norm": 1.8319112062454224,
|
| 848 |
+
"learning_rate": 6.366115251772405e-06,
|
| 849 |
+
"loss": 1.6403,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.3666111195273443,
|
| 854 |
+
"grad_norm": 1.6500194072723389,
|
| 855 |
+
"learning_rate": 6.335817730109678e-06,
|
| 856 |
+
"loss": 1.6216,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.3696409634903802,
|
| 861 |
+
"grad_norm": 1.6576268672943115,
|
| 862 |
+
"learning_rate": 6.305520208446949e-06,
|
| 863 |
+
"loss": 1.6326,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.37267080745341613,
|
| 868 |
+
"grad_norm": 2.2077317237854004,
|
| 869 |
+
"learning_rate": 6.275222686784221e-06,
|
| 870 |
+
"loss": 1.609,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.37570065141645204,
|
| 875 |
+
"grad_norm": 1.4579232931137085,
|
| 876 |
+
"learning_rate": 6.244925165121494e-06,
|
| 877 |
+
"loss": 1.5816,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.37873049537948794,
|
| 882 |
+
"grad_norm": 1.6647648811340332,
|
| 883 |
+
"learning_rate": 6.214627643458766e-06,
|
| 884 |
+
"loss": 1.6208,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.38176033934252385,
|
| 889 |
+
"grad_norm": 1.413583755493164,
|
| 890 |
+
"learning_rate": 6.184330121796037e-06,
|
| 891 |
+
"loss": 1.6206,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.38479018330555975,
|
| 896 |
+
"grad_norm": 1.743116021156311,
|
| 897 |
+
"learning_rate": 6.15403260013331e-06,
|
| 898 |
+
"loss": 1.5973,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.38782002726859566,
|
| 903 |
+
"grad_norm": 1.8276968002319336,
|
| 904 |
+
"learning_rate": 6.123735078470582e-06,
|
| 905 |
+
"loss": 1.6384,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.39084987123163156,
|
| 910 |
+
"grad_norm": 1.3672616481781006,
|
| 911 |
+
"learning_rate": 6.093437556807854e-06,
|
| 912 |
+
"loss": 1.5883,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.39387971519466747,
|
| 917 |
+
"grad_norm": 1.6729201078414917,
|
| 918 |
+
"learning_rate": 6.063140035145126e-06,
|
| 919 |
+
"loss": 1.6074,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.3969095591577034,
|
| 924 |
+
"grad_norm": 1.433160662651062,
|
| 925 |
+
"learning_rate": 6.0328425134823976e-06,
|
| 926 |
+
"loss": 1.6359,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.3999394031207393,
|
| 931 |
+
"grad_norm": 1.6422028541564941,
|
| 932 |
+
"learning_rate": 6.0025449918196695e-06,
|
| 933 |
+
"loss": 1.6121,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.4029692470837752,
|
| 938 |
+
"grad_norm": 1.6269747018814087,
|
| 939 |
+
"learning_rate": 5.972247470156942e-06,
|
| 940 |
+
"loss": 1.6143,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.4059990910468111,
|
| 945 |
+
"grad_norm": 1.5140916109085083,
|
| 946 |
+
"learning_rate": 5.941949948494213e-06,
|
| 947 |
+
"loss": 1.5386,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.409028935009847,
|
| 952 |
+
"grad_norm": 1.6418771743774414,
|
| 953 |
+
"learning_rate": 5.911652426831485e-06,
|
| 954 |
+
"loss": 1.5685,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.4120587789728829,
|
| 959 |
+
"grad_norm": 2.016139268875122,
|
| 960 |
+
"learning_rate": 5.881354905168758e-06,
|
| 961 |
+
"loss": 1.6092,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.4150886229359188,
|
| 966 |
+
"grad_norm": 1.8047841787338257,
|
| 967 |
+
"learning_rate": 5.851057383506029e-06,
|
| 968 |
+
"loss": 1.6405,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.4181184668989547,
|
| 973 |
+
"grad_norm": 1.426552653312683,
|
| 974 |
+
"learning_rate": 5.820759861843301e-06,
|
| 975 |
+
"loss": 1.5648,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.4211483108619906,
|
| 980 |
+
"grad_norm": 1.8354159593582153,
|
| 981 |
+
"learning_rate": 5.790462340180574e-06,
|
| 982 |
+
"loss": 1.6048,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.4241781548250265,
|
| 987 |
+
"grad_norm": 1.475692868232727,
|
| 988 |
+
"learning_rate": 5.760164818517846e-06,
|
| 989 |
+
"loss": 1.5869,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.4272079987880624,
|
| 994 |
+
"grad_norm": 1.6546450853347778,
|
| 995 |
+
"learning_rate": 5.729867296855117e-06,
|
| 996 |
+
"loss": 1.5866,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.43023784275109833,
|
| 1001 |
+
"grad_norm": 1.5041762590408325,
|
| 1002 |
+
"learning_rate": 5.69956977519239e-06,
|
| 1003 |
+
"loss": 1.6071,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.43326768671413424,
|
| 1008 |
+
"grad_norm": 1.5460017919540405,
|
| 1009 |
+
"learning_rate": 5.669272253529662e-06,
|
| 1010 |
+
"loss": 1.5823,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.43629753067717014,
|
| 1015 |
+
"grad_norm": 1.6852233409881592,
|
| 1016 |
+
"learning_rate": 5.6389747318669345e-06,
|
| 1017 |
+
"loss": 1.6109,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.43932737464020605,
|
| 1022 |
+
"grad_norm": 1.5145368576049805,
|
| 1023 |
+
"learning_rate": 5.6086772102042056e-06,
|
| 1024 |
+
"loss": 1.5887,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.44235721860324195,
|
| 1029 |
+
"grad_norm": 1.6494066715240479,
|
| 1030 |
+
"learning_rate": 5.5783796885414775e-06,
|
| 1031 |
+
"loss": 1.5881,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.44538706256627786,
|
| 1036 |
+
"grad_norm": 1.7732048034667969,
|
| 1037 |
+
"learning_rate": 5.54808216687875e-06,
|
| 1038 |
+
"loss": 1.5923,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.44841690652931376,
|
| 1043 |
+
"grad_norm": 2.078328847885132,
|
| 1044 |
+
"learning_rate": 5.517784645216022e-06,
|
| 1045 |
+
"loss": 1.6117,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.45144675049234967,
|
| 1050 |
+
"grad_norm": 1.5237518548965454,
|
| 1051 |
+
"learning_rate": 5.487487123553293e-06,
|
| 1052 |
+
"loss": 1.6206,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.4544765944553856,
|
| 1057 |
+
"grad_norm": 1.5857949256896973,
|
| 1058 |
+
"learning_rate": 5.457189601890566e-06,
|
| 1059 |
+
"loss": 1.5938,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.4575064384184214,
|
| 1064 |
+
"grad_norm": 1.8254096508026123,
|
| 1065 |
+
"learning_rate": 5.426892080227838e-06,
|
| 1066 |
+
"loss": 1.6217,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.46053628238145733,
|
| 1071 |
+
"grad_norm": 1.7777276039123535,
|
| 1072 |
+
"learning_rate": 5.39659455856511e-06,
|
| 1073 |
+
"loss": 1.5695,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.46356612634449323,
|
| 1078 |
+
"grad_norm": 1.6662938594818115,
|
| 1079 |
+
"learning_rate": 5.366297036902381e-06,
|
| 1080 |
+
"loss": 1.5875,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.46659597030752914,
|
| 1085 |
+
"grad_norm": 1.6636159420013428,
|
| 1086 |
+
"learning_rate": 5.335999515239654e-06,
|
| 1087 |
+
"loss": 1.5969,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.46962581427056505,
|
| 1092 |
+
"grad_norm": 1.4926279783248901,
|
| 1093 |
+
"learning_rate": 5.305701993576926e-06,
|
| 1094 |
+
"loss": 1.5905,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.47265565823360095,
|
| 1099 |
+
"grad_norm": 1.566446304321289,
|
| 1100 |
+
"learning_rate": 5.275404471914199e-06,
|
| 1101 |
+
"loss": 1.6126,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.47568550219663686,
|
| 1106 |
+
"grad_norm": 1.8684853315353394,
|
| 1107 |
+
"learning_rate": 5.24510695025147e-06,
|
| 1108 |
+
"loss": 1.5737,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.47871534615967276,
|
| 1113 |
+
"grad_norm": 1.8654356002807617,
|
| 1114 |
+
"learning_rate": 5.214809428588742e-06,
|
| 1115 |
+
"loss": 1.5964,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.48174519012270867,
|
| 1120 |
+
"grad_norm": 1.3818305730819702,
|
| 1121 |
+
"learning_rate": 5.184511906926014e-06,
|
| 1122 |
+
"loss": 1.5757,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.48477503408574457,
|
| 1127 |
+
"grad_norm": 1.6214017868041992,
|
| 1128 |
+
"learning_rate": 5.154214385263286e-06,
|
| 1129 |
+
"loss": 1.559,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.4878048780487805,
|
| 1134 |
+
"grad_norm": 1.5026991367340088,
|
| 1135 |
+
"learning_rate": 5.1239168636005574e-06,
|
| 1136 |
+
"loss": 1.6033,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.4908347220118164,
|
| 1141 |
+
"grad_norm": 1.4133880138397217,
|
| 1142 |
+
"learning_rate": 5.09361934193783e-06,
|
| 1143 |
+
"loss": 1.5947,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.4938645659748523,
|
| 1148 |
+
"grad_norm": 1.5185195207595825,
|
| 1149 |
+
"learning_rate": 5.063321820275102e-06,
|
| 1150 |
+
"loss": 1.5858,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.4968944099378882,
|
| 1155 |
+
"grad_norm": 1.5284918546676636,
|
| 1156 |
+
"learning_rate": 5.033024298612373e-06,
|
| 1157 |
+
"loss": 1.5998,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.4999242539009241,
|
| 1162 |
+
"grad_norm": 1.5977320671081543,
|
| 1163 |
+
"learning_rate": 5.002726776949646e-06,
|
| 1164 |
+
"loss": 1.5609,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.5029540978639601,
|
| 1169 |
+
"grad_norm": 1.6577131748199463,
|
| 1170 |
+
"learning_rate": 4.972429255286918e-06,
|
| 1171 |
+
"loss": 1.5684,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.5059839418269959,
|
| 1176 |
+
"grad_norm": 1.5592504739761353,
|
| 1177 |
+
"learning_rate": 4.94213173362419e-06,
|
| 1178 |
+
"loss": 1.597,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.5090137857900318,
|
| 1183 |
+
"grad_norm": 1.6733429431915283,
|
| 1184 |
+
"learning_rate": 4.911834211961462e-06,
|
| 1185 |
+
"loss": 1.5803,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.5120436297530677,
|
| 1190 |
+
"grad_norm": 1.390070915222168,
|
| 1191 |
+
"learning_rate": 4.881536690298734e-06,
|
| 1192 |
+
"loss": 1.5868,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.5150734737161036,
|
| 1197 |
+
"grad_norm": 1.6423391103744507,
|
| 1198 |
+
"learning_rate": 4.851239168636006e-06,
|
| 1199 |
+
"loss": 1.6197,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.5181033176791395,
|
| 1204 |
+
"grad_norm": 1.5211403369903564,
|
| 1205 |
+
"learning_rate": 4.820941646973278e-06,
|
| 1206 |
+
"loss": 1.5985,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.5211331616421754,
|
| 1211 |
+
"grad_norm": 1.3367249965667725,
|
| 1212 |
+
"learning_rate": 4.7906441253105504e-06,
|
| 1213 |
+
"loss": 1.563,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.5241630056052113,
|
| 1218 |
+
"grad_norm": 1.4820023775100708,
|
| 1219 |
+
"learning_rate": 4.7603466036478215e-06,
|
| 1220 |
+
"loss": 1.5519,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.5271928495682472,
|
| 1225 |
+
"grad_norm": 1.5887796878814697,
|
| 1226 |
+
"learning_rate": 4.7306550324183485e-06,
|
| 1227 |
+
"loss": 1.5901,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.5302226935312832,
|
| 1232 |
+
"grad_norm": 2.0754947662353516,
|
| 1233 |
+
"learning_rate": 4.7003575107556205e-06,
|
| 1234 |
+
"loss": 1.58,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.533252537494319,
|
| 1239 |
+
"grad_norm": 1.545721173286438,
|
| 1240 |
+
"learning_rate": 4.670059989092892e-06,
|
| 1241 |
+
"loss": 1.5588,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.536282381457355,
|
| 1246 |
+
"grad_norm": 1.69772469997406,
|
| 1247 |
+
"learning_rate": 4.639762467430164e-06,
|
| 1248 |
+
"loss": 1.5847,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.5393122254203908,
|
| 1253 |
+
"grad_norm": 1.5981988906860352,
|
| 1254 |
+
"learning_rate": 4.609464945767437e-06,
|
| 1255 |
+
"loss": 1.5508,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.5423420693834268,
|
| 1260 |
+
"grad_norm": 1.884867787361145,
|
| 1261 |
+
"learning_rate": 4.579167424104708e-06,
|
| 1262 |
+
"loss": 1.5583,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.5453719133464626,
|
| 1267 |
+
"grad_norm": 1.8997677564620972,
|
| 1268 |
+
"learning_rate": 4.548869902441981e-06,
|
| 1269 |
+
"loss": 1.5681,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.5484017573094986,
|
| 1274 |
+
"grad_norm": 1.6099635362625122,
|
| 1275 |
+
"learning_rate": 4.518572380779253e-06,
|
| 1276 |
+
"loss": 1.5838,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.5514316012725344,
|
| 1281 |
+
"grad_norm": 1.659430742263794,
|
| 1282 |
+
"learning_rate": 4.488274859116525e-06,
|
| 1283 |
+
"loss": 1.5876,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.5544614452355704,
|
| 1288 |
+
"grad_norm": 1.72199547290802,
|
| 1289 |
+
"learning_rate": 4.457977337453797e-06,
|
| 1290 |
+
"loss": 1.5776,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.5574912891986062,
|
| 1295 |
+
"grad_norm": 1.5597950220108032,
|
| 1296 |
+
"learning_rate": 4.427679815791069e-06,
|
| 1297 |
+
"loss": 1.5383,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.5605211331616422,
|
| 1302 |
+
"grad_norm": 1.8483946323394775,
|
| 1303 |
+
"learning_rate": 4.397382294128341e-06,
|
| 1304 |
+
"loss": 1.6229,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.563550977124678,
|
| 1309 |
+
"grad_norm": 1.459078311920166,
|
| 1310 |
+
"learning_rate": 4.367084772465613e-06,
|
| 1311 |
+
"loss": 1.5796,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.566580821087714,
|
| 1316 |
+
"grad_norm": 1.5624710321426392,
|
| 1317 |
+
"learning_rate": 4.336787250802885e-06,
|
| 1318 |
+
"loss": 1.5648,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.5696106650507499,
|
| 1323 |
+
"grad_norm": 1.5045188665390015,
|
| 1324 |
+
"learning_rate": 4.306489729140157e-06,
|
| 1325 |
+
"loss": 1.6078,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.5726405090137858,
|
| 1330 |
+
"grad_norm": 1.5284470319747925,
|
| 1331 |
+
"learning_rate": 4.2761922074774285e-06,
|
| 1332 |
+
"loss": 1.5581,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.5756703529768217,
|
| 1337 |
+
"grad_norm": 1.7991974353790283,
|
| 1338 |
+
"learning_rate": 4.2458946858147e-06,
|
| 1339 |
+
"loss": 1.5629,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.5787001969398576,
|
| 1344 |
+
"grad_norm": 1.5898367166519165,
|
| 1345 |
+
"learning_rate": 4.215597164151972e-06,
|
| 1346 |
+
"loss": 1.6102,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.5817300409028935,
|
| 1351 |
+
"grad_norm": 1.5055047273635864,
|
| 1352 |
+
"learning_rate": 4.185299642489244e-06,
|
| 1353 |
+
"loss": 1.5602,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.5847598848659294,
|
| 1358 |
+
"grad_norm": 1.4318591356277466,
|
| 1359 |
+
"learning_rate": 4.155002120826517e-06,
|
| 1360 |
+
"loss": 1.5558,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.5877897288289653,
|
| 1365 |
+
"grad_norm": 1.6692588329315186,
|
| 1366 |
+
"learning_rate": 4.124704599163788e-06,
|
| 1367 |
+
"loss": 1.573,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.5908195727920013,
|
| 1372 |
+
"grad_norm": 1.4996757507324219,
|
| 1373 |
+
"learning_rate": 4.094407077501061e-06,
|
| 1374 |
+
"loss": 1.5641,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.5938494167550371,
|
| 1379 |
+
"grad_norm": 1.491071343421936,
|
| 1380 |
+
"learning_rate": 4.064109555838333e-06,
|
| 1381 |
+
"loss": 1.5546,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.5968792607180731,
|
| 1386 |
+
"grad_norm": 1.5985503196716309,
|
| 1387 |
+
"learning_rate": 4.033812034175605e-06,
|
| 1388 |
+
"loss": 1.5443,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.5999091046811089,
|
| 1393 |
+
"grad_norm": 1.6403000354766846,
|
| 1394 |
+
"learning_rate": 4.003514512512877e-06,
|
| 1395 |
+
"loss": 1.591,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.6029389486441449,
|
| 1400 |
+
"grad_norm": 1.3770664930343628,
|
| 1401 |
+
"learning_rate": 3.973216990850149e-06,
|
| 1402 |
+
"loss": 1.5142,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.6059687926071807,
|
| 1407 |
+
"grad_norm": 1.6218574047088623,
|
| 1408 |
+
"learning_rate": 3.942919469187421e-06,
|
| 1409 |
+
"loss": 1.5854,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.6089986365702167,
|
| 1414 |
+
"grad_norm": 1.2990355491638184,
|
| 1415 |
+
"learning_rate": 3.9126219475246926e-06,
|
| 1416 |
+
"loss": 1.5595,
|
| 1417 |
+
"step": 10050
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.6120284805332525,
|
| 1421 |
+
"grad_norm": 1.4130761623382568,
|
| 1422 |
+
"learning_rate": 3.8823244258619645e-06,
|
| 1423 |
+
"loss": 1.5471,
|
| 1424 |
+
"step": 10100
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.6150583244962884,
|
| 1428 |
+
"grad_norm": 1.4140211343765259,
|
| 1429 |
+
"learning_rate": 3.852026904199237e-06,
|
| 1430 |
+
"loss": 1.5071,
|
| 1431 |
+
"step": 10150
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.6180881684593243,
|
| 1435 |
+
"grad_norm": 1.4824833869934082,
|
| 1436 |
+
"learning_rate": 3.821729382536508e-06,
|
| 1437 |
+
"loss": 1.5711,
|
| 1438 |
+
"step": 10200
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.6211180124223602,
|
| 1442 |
+
"grad_norm": 1.6242071390151978,
|
| 1443 |
+
"learning_rate": 3.791431860873781e-06,
|
| 1444 |
+
"loss": 1.5181,
|
| 1445 |
+
"step": 10250
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.6241478563853962,
|
| 1449 |
+
"grad_norm": 1.4817885160446167,
|
| 1450 |
+
"learning_rate": 3.7611343392110527e-06,
|
| 1451 |
+
"loss": 1.5468,
|
| 1452 |
+
"step": 10300
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.627177700348432,
|
| 1456 |
+
"grad_norm": 1.3601789474487305,
|
| 1457 |
+
"learning_rate": 3.730836817548325e-06,
|
| 1458 |
+
"loss": 1.5761,
|
| 1459 |
+
"step": 10350
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.630207544311468,
|
| 1463 |
+
"grad_norm": 1.3270106315612793,
|
| 1464 |
+
"learning_rate": 3.700539295885597e-06,
|
| 1465 |
+
"loss": 1.5336,
|
| 1466 |
+
"step": 10400
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.6332373882745038,
|
| 1470 |
+
"grad_norm": 1.4662185907363892,
|
| 1471 |
+
"learning_rate": 3.670241774222869e-06,
|
| 1472 |
+
"loss": 1.5958,
|
| 1473 |
+
"step": 10450
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.6362672322375398,
|
| 1477 |
+
"grad_norm": 1.480850100517273,
|
| 1478 |
+
"learning_rate": 3.639944252560141e-06,
|
| 1479 |
+
"loss": 1.565,
|
| 1480 |
+
"step": 10500
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.6392970762005756,
|
| 1484 |
+
"grad_norm": 1.4275442361831665,
|
| 1485 |
+
"learning_rate": 3.6096467308974132e-06,
|
| 1486 |
+
"loss": 1.5316,
|
| 1487 |
+
"step": 10550
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.6423269201636116,
|
| 1491 |
+
"grad_norm": 1.709631323814392,
|
| 1492 |
+
"learning_rate": 3.5793492092346847e-06,
|
| 1493 |
+
"loss": 1.5568,
|
| 1494 |
+
"step": 10600
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.6453567641266474,
|
| 1498 |
+
"grad_norm": 1.4881377220153809,
|
| 1499 |
+
"learning_rate": 3.549051687571957e-06,
|
| 1500 |
+
"loss": 1.5452,
|
| 1501 |
+
"step": 10650
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.6483866080896834,
|
| 1505 |
+
"grad_norm": 1.3687924146652222,
|
| 1506 |
+
"learning_rate": 3.518754165909229e-06,
|
| 1507 |
+
"loss": 1.567,
|
| 1508 |
+
"step": 10700
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.6514164520527193,
|
| 1512 |
+
"grad_norm": 1.2314640283584595,
|
| 1513 |
+
"learning_rate": 3.4884566442465006e-06,
|
| 1514 |
+
"loss": 1.5678,
|
| 1515 |
+
"step": 10750
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.6544462960157552,
|
| 1519 |
+
"grad_norm": 1.5389395952224731,
|
| 1520 |
+
"learning_rate": 3.458159122583773e-06,
|
| 1521 |
+
"loss": 1.5624,
|
| 1522 |
+
"step": 10800
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.6574761399787911,
|
| 1526 |
+
"grad_norm": 1.401017189025879,
|
| 1527 |
+
"learning_rate": 3.427861600921045e-06,
|
| 1528 |
+
"loss": 1.5287,
|
| 1529 |
+
"step": 10850
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.660505983941827,
|
| 1533 |
+
"grad_norm": 1.9174494743347168,
|
| 1534 |
+
"learning_rate": 3.397564079258317e-06,
|
| 1535 |
+
"loss": 1.547,
|
| 1536 |
+
"step": 10900
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.6635358279048629,
|
| 1540 |
+
"grad_norm": 1.5186442136764526,
|
| 1541 |
+
"learning_rate": 3.3672665575955887e-06,
|
| 1542 |
+
"loss": 1.5489,
|
| 1543 |
+
"step": 10950
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.6665656718678988,
|
| 1547 |
+
"grad_norm": 1.3896212577819824,
|
| 1548 |
+
"learning_rate": 3.336969035932861e-06,
|
| 1549 |
+
"loss": 1.5368,
|
| 1550 |
+
"step": 11000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.6695955158309347,
|
| 1554 |
+
"grad_norm": 1.7711433172225952,
|
| 1555 |
+
"learning_rate": 3.3072774647033872e-06,
|
| 1556 |
+
"loss": 1.5129,
|
| 1557 |
+
"step": 11050
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.6726253597939706,
|
| 1561 |
+
"grad_norm": 1.5082119703292847,
|
| 1562 |
+
"learning_rate": 3.2769799430406596e-06,
|
| 1563 |
+
"loss": 1.5491,
|
| 1564 |
+
"step": 11100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.6756552037570065,
|
| 1568 |
+
"grad_norm": 1.3804906606674194,
|
| 1569 |
+
"learning_rate": 3.246682421377931e-06,
|
| 1570 |
+
"loss": 1.5687,
|
| 1571 |
+
"step": 11150
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.6786850477200425,
|
| 1575 |
+
"grad_norm": 1.6013615131378174,
|
| 1576 |
+
"learning_rate": 3.2163848997152035e-06,
|
| 1577 |
+
"loss": 1.5891,
|
| 1578 |
+
"step": 11200
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.6817148916830783,
|
| 1582 |
+
"grad_norm": 1.6114518642425537,
|
| 1583 |
+
"learning_rate": 3.1860873780524754e-06,
|
| 1584 |
+
"loss": 1.5891,
|
| 1585 |
+
"step": 11250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.6847447356461143,
|
| 1589 |
+
"grad_norm": 1.5464235544204712,
|
| 1590 |
+
"learning_rate": 3.1557898563897478e-06,
|
| 1591 |
+
"loss": 1.5576,
|
| 1592 |
+
"step": 11300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.6877745796091501,
|
| 1596 |
+
"grad_norm": 1.3998115062713623,
|
| 1597 |
+
"learning_rate": 3.1254923347270193e-06,
|
| 1598 |
+
"loss": 1.543,
|
| 1599 |
+
"step": 11350
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.6908044235721861,
|
| 1603 |
+
"grad_norm": 1.5496994256973267,
|
| 1604 |
+
"learning_rate": 3.0951948130642917e-06,
|
| 1605 |
+
"loss": 1.5751,
|
| 1606 |
+
"step": 11400
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.6938342675352219,
|
| 1610 |
+
"grad_norm": 1.4767708778381348,
|
| 1611 |
+
"learning_rate": 3.0648972914015636e-06,
|
| 1612 |
+
"loss": 1.5892,
|
| 1613 |
+
"step": 11450
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.6968641114982579,
|
| 1617 |
+
"grad_norm": 1.5664912462234497,
|
| 1618 |
+
"learning_rate": 3.0345997697388355e-06,
|
| 1619 |
+
"loss": 1.5674,
|
| 1620 |
+
"step": 11500
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.6998939554612937,
|
| 1624 |
+
"grad_norm": 1.5581713914871216,
|
| 1625 |
+
"learning_rate": 3.0043022480761075e-06,
|
| 1626 |
+
"loss": 1.5157,
|
| 1627 |
+
"step": 11550
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.7029237994243297,
|
| 1631 |
+
"grad_norm": 1.4770848751068115,
|
| 1632 |
+
"learning_rate": 2.97400472641338e-06,
|
| 1633 |
+
"loss": 1.5241,
|
| 1634 |
+
"step": 11600
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.7059536433873655,
|
| 1638 |
+
"grad_norm": 1.5183898210525513,
|
| 1639 |
+
"learning_rate": 2.9437072047506514e-06,
|
| 1640 |
+
"loss": 1.5752,
|
| 1641 |
+
"step": 11650
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.7089834873504015,
|
| 1645 |
+
"grad_norm": 1.3331156969070435,
|
| 1646 |
+
"learning_rate": 2.9134096830879237e-06,
|
| 1647 |
+
"loss": 1.5524,
|
| 1648 |
+
"step": 11700
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.7120133313134374,
|
| 1652 |
+
"grad_norm": 1.3551660776138306,
|
| 1653 |
+
"learning_rate": 2.8831121614251957e-06,
|
| 1654 |
+
"loss": 1.5653,
|
| 1655 |
+
"step": 11750
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.7150431752764732,
|
| 1659 |
+
"grad_norm": 1.4175201654434204,
|
| 1660 |
+
"learning_rate": 2.852814639762468e-06,
|
| 1661 |
+
"loss": 1.5538,
|
| 1662 |
+
"step": 11800
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.7180730192395092,
|
| 1666 |
+
"grad_norm": 1.482858657836914,
|
| 1667 |
+
"learning_rate": 2.8225171180997395e-06,
|
| 1668 |
+
"loss": 1.5411,
|
| 1669 |
+
"step": 11850
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.721102863202545,
|
| 1673 |
+
"grad_norm": 1.3337117433547974,
|
| 1674 |
+
"learning_rate": 2.792219596437012e-06,
|
| 1675 |
+
"loss": 1.5402,
|
| 1676 |
+
"step": 11900
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.724132707165581,
|
| 1680 |
+
"grad_norm": 1.529326319694519,
|
| 1681 |
+
"learning_rate": 2.761922074774284e-06,
|
| 1682 |
+
"loss": 1.5357,
|
| 1683 |
+
"step": 11950
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.7271625511286168,
|
| 1687 |
+
"grad_norm": 1.40908944606781,
|
| 1688 |
+
"learning_rate": 2.7316245531115558e-06,
|
| 1689 |
+
"loss": 1.521,
|
| 1690 |
+
"step": 12000
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.7301923950916528,
|
| 1694 |
+
"grad_norm": 1.434410572052002,
|
| 1695 |
+
"learning_rate": 2.7013270314488277e-06,
|
| 1696 |
+
"loss": 1.5723,
|
| 1697 |
+
"step": 12050
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.7332222390546886,
|
| 1701 |
+
"grad_norm": 1.5077331066131592,
|
| 1702 |
+
"learning_rate": 2.6710295097861e-06,
|
| 1703 |
+
"loss": 1.5979,
|
| 1704 |
+
"step": 12100
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.7362520830177246,
|
| 1708 |
+
"grad_norm": 1.4637471437454224,
|
| 1709 |
+
"learning_rate": 2.6407319881233716e-06,
|
| 1710 |
+
"loss": 1.5465,
|
| 1711 |
+
"step": 12150
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.7392819269807605,
|
| 1715 |
+
"grad_norm": 1.7819764614105225,
|
| 1716 |
+
"learning_rate": 2.6104344664606435e-06,
|
| 1717 |
+
"loss": 1.5425,
|
| 1718 |
+
"step": 12200
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.7423117709437964,
|
| 1722 |
+
"grad_norm": 1.3142591714859009,
|
| 1723 |
+
"learning_rate": 2.580136944797916e-06,
|
| 1724 |
+
"loss": 1.5659,
|
| 1725 |
+
"step": 12250
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.7453416149068323,
|
| 1729 |
+
"grad_norm": 1.4326512813568115,
|
| 1730 |
+
"learning_rate": 2.5498394231351874e-06,
|
| 1731 |
+
"loss": 1.5506,
|
| 1732 |
+
"step": 12300
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.7483714588698682,
|
| 1736 |
+
"grad_norm": 1.3926914930343628,
|
| 1737 |
+
"learning_rate": 2.5195419014724598e-06,
|
| 1738 |
+
"loss": 1.519,
|
| 1739 |
+
"step": 12350
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.7514013028329041,
|
| 1743 |
+
"grad_norm": 1.5886470079421997,
|
| 1744 |
+
"learning_rate": 2.4892443798097317e-06,
|
| 1745 |
+
"loss": 1.5323,
|
| 1746 |
+
"step": 12400
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.75443114679594,
|
| 1750 |
+
"grad_norm": 1.6596341133117676,
|
| 1751 |
+
"learning_rate": 2.4589468581470036e-06,
|
| 1752 |
+
"loss": 1.5572,
|
| 1753 |
+
"step": 12450
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.7574609907589759,
|
| 1757 |
+
"grad_norm": 1.5624167919158936,
|
| 1758 |
+
"learning_rate": 2.428649336484276e-06,
|
| 1759 |
+
"loss": 1.517,
|
| 1760 |
+
"step": 12500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.7604908347220118,
|
| 1764 |
+
"grad_norm": 1.5052602291107178,
|
| 1765 |
+
"learning_rate": 2.398351814821548e-06,
|
| 1766 |
+
"loss": 1.5634,
|
| 1767 |
+
"step": 12550
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.7635206786850477,
|
| 1771 |
+
"grad_norm": 1.4568336009979248,
|
| 1772 |
+
"learning_rate": 2.36805429315882e-06,
|
| 1773 |
+
"loss": 1.5581,
|
| 1774 |
+
"step": 12600
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.7665505226480837,
|
| 1778 |
+
"grad_norm": 1.4686074256896973,
|
| 1779 |
+
"learning_rate": 2.337756771496092e-06,
|
| 1780 |
+
"loss": 1.5383,
|
| 1781 |
+
"step": 12650
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.7695803666111195,
|
| 1785 |
+
"grad_norm": 1.53707754611969,
|
| 1786 |
+
"learning_rate": 2.3074592498333638e-06,
|
| 1787 |
+
"loss": 1.5654,
|
| 1788 |
+
"step": 12700
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.7726102105741555,
|
| 1792 |
+
"grad_norm": 1.5452001094818115,
|
| 1793 |
+
"learning_rate": 2.277161728170636e-06,
|
| 1794 |
+
"loss": 1.5772,
|
| 1795 |
+
"step": 12750
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.7756400545371913,
|
| 1799 |
+
"grad_norm": 1.553062915802002,
|
| 1800 |
+
"learning_rate": 2.2468642065079076e-06,
|
| 1801 |
+
"loss": 1.5178,
|
| 1802 |
+
"step": 12800
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.7786698985002273,
|
| 1806 |
+
"grad_norm": 1.328934669494629,
|
| 1807 |
+
"learning_rate": 2.2165666848451796e-06,
|
| 1808 |
+
"loss": 1.5718,
|
| 1809 |
+
"step": 12850
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.7816997424632631,
|
| 1813 |
+
"grad_norm": 1.7127735614776611,
|
| 1814 |
+
"learning_rate": 2.186269163182452e-06,
|
| 1815 |
+
"loss": 1.5341,
|
| 1816 |
+
"step": 12900
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.7847295864262991,
|
| 1820 |
+
"grad_norm": 1.728936791419983,
|
| 1821 |
+
"learning_rate": 2.155971641519724e-06,
|
| 1822 |
+
"loss": 1.5546,
|
| 1823 |
+
"step": 12950
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.7877594303893349,
|
| 1827 |
+
"grad_norm": 1.6442391872406006,
|
| 1828 |
+
"learning_rate": 2.125674119856996e-06,
|
| 1829 |
+
"loss": 1.56,
|
| 1830 |
+
"step": 13000
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.7907892743523709,
|
| 1834 |
+
"grad_norm": 1.4701004028320312,
|
| 1835 |
+
"learning_rate": 2.0953765981942678e-06,
|
| 1836 |
+
"loss": 1.5182,
|
| 1837 |
+
"step": 13050
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.7938191183154067,
|
| 1841 |
+
"grad_norm": 1.2920253276824951,
|
| 1842 |
+
"learning_rate": 2.0650790765315397e-06,
|
| 1843 |
+
"loss": 1.5291,
|
| 1844 |
+
"step": 13100
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.7968489622784427,
|
| 1848 |
+
"grad_norm": 1.3979644775390625,
|
| 1849 |
+
"learning_rate": 2.034781554868812e-06,
|
| 1850 |
+
"loss": 1.5246,
|
| 1851 |
+
"step": 13150
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.7998788062414786,
|
| 1855 |
+
"grad_norm": 1.7545385360717773,
|
| 1856 |
+
"learning_rate": 2.004484033206084e-06,
|
| 1857 |
+
"loss": 1.5752,
|
| 1858 |
+
"step": 13200
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.8029086502045145,
|
| 1862 |
+
"grad_norm": 1.347290277481079,
|
| 1863 |
+
"learning_rate": 1.974186511543356e-06,
|
| 1864 |
+
"loss": 1.5264,
|
| 1865 |
+
"step": 13250
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.8059384941675504,
|
| 1869 |
+
"grad_norm": 1.4548012018203735,
|
| 1870 |
+
"learning_rate": 1.943888989880628e-06,
|
| 1871 |
+
"loss": 1.5995,
|
| 1872 |
+
"step": 13300
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.8089683381305863,
|
| 1876 |
+
"grad_norm": 1.6475197076797485,
|
| 1877 |
+
"learning_rate": 1.9135914682179e-06,
|
| 1878 |
+
"loss": 1.5506,
|
| 1879 |
+
"step": 13350
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.8119981820936222,
|
| 1883 |
+
"grad_norm": 1.497528314590454,
|
| 1884 |
+
"learning_rate": 1.883293946555172e-06,
|
| 1885 |
+
"loss": 1.5702,
|
| 1886 |
+
"step": 13400
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.815028026056658,
|
| 1890 |
+
"grad_norm": 1.791481852531433,
|
| 1891 |
+
"learning_rate": 1.852996424892444e-06,
|
| 1892 |
+
"loss": 1.5474,
|
| 1893 |
+
"step": 13450
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.818057870019694,
|
| 1897 |
+
"grad_norm": 1.5969312191009521,
|
| 1898 |
+
"learning_rate": 1.822698903229716e-06,
|
| 1899 |
+
"loss": 1.5416,
|
| 1900 |
+
"step": 13500
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.8210877139827298,
|
| 1904 |
+
"grad_norm": 1.4239813089370728,
|
| 1905 |
+
"learning_rate": 1.792401381566988e-06,
|
| 1906 |
+
"loss": 1.5252,
|
| 1907 |
+
"step": 13550
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.8241175579457658,
|
| 1911 |
+
"grad_norm": 1.579437017440796,
|
| 1912 |
+
"learning_rate": 1.76210385990426e-06,
|
| 1913 |
+
"loss": 1.5764,
|
| 1914 |
+
"step": 13600
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.8271474019088016,
|
| 1918 |
+
"grad_norm": 1.6514719724655151,
|
| 1919 |
+
"learning_rate": 1.731806338241532e-06,
|
| 1920 |
+
"loss": 1.545,
|
| 1921 |
+
"step": 13650
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.8301772458718376,
|
| 1925 |
+
"grad_norm": 1.5468313694000244,
|
| 1926 |
+
"learning_rate": 1.701508816578804e-06,
|
| 1927 |
+
"loss": 1.5377,
|
| 1928 |
+
"step": 13700
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.8332070898348735,
|
| 1932 |
+
"grad_norm": 1.8032771348953247,
|
| 1933 |
+
"learning_rate": 1.6712112949160762e-06,
|
| 1934 |
+
"loss": 1.5183,
|
| 1935 |
+
"step": 13750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.8362369337979094,
|
| 1939 |
+
"grad_norm": 2.1406657695770264,
|
| 1940 |
+
"learning_rate": 1.6409137732533481e-06,
|
| 1941 |
+
"loss": 1.5268,
|
| 1942 |
+
"step": 13800
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.8392667777609453,
|
| 1946 |
+
"grad_norm": 1.7589893341064453,
|
| 1947 |
+
"learning_rate": 1.61061625159062e-06,
|
| 1948 |
+
"loss": 1.5054,
|
| 1949 |
+
"step": 13850
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.8422966217239812,
|
| 1953 |
+
"grad_norm": 1.619626522064209,
|
| 1954 |
+
"learning_rate": 1.5803187299278922e-06,
|
| 1955 |
+
"loss": 1.5747,
|
| 1956 |
+
"step": 13900
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.8453264656870171,
|
| 1960 |
+
"grad_norm": 1.3544018268585205,
|
| 1961 |
+
"learning_rate": 1.5500212082651641e-06,
|
| 1962 |
+
"loss": 1.5264,
|
| 1963 |
+
"step": 13950
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.848356309650053,
|
| 1967 |
+
"grad_norm": 1.3174818754196167,
|
| 1968 |
+
"learning_rate": 1.5197236866024359e-06,
|
| 1969 |
+
"loss": 1.4991,
|
| 1970 |
+
"step": 14000
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.8513861536130889,
|
| 1974 |
+
"grad_norm": 1.3408424854278564,
|
| 1975 |
+
"learning_rate": 1.489426164939708e-06,
|
| 1976 |
+
"loss": 1.5574,
|
| 1977 |
+
"step": 14050
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.8544159975761249,
|
| 1981 |
+
"grad_norm": 1.5097547769546509,
|
| 1982 |
+
"learning_rate": 1.45912864327698e-06,
|
| 1983 |
+
"loss": 1.5423,
|
| 1984 |
+
"step": 14100
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.8574458415391607,
|
| 1988 |
+
"grad_norm": 1.5766481161117554,
|
| 1989 |
+
"learning_rate": 1.4288311216142519e-06,
|
| 1990 |
+
"loss": 1.5253,
|
| 1991 |
+
"step": 14150
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.8604756855021967,
|
| 1995 |
+
"grad_norm": 1.2671705484390259,
|
| 1996 |
+
"learning_rate": 1.398533599951524e-06,
|
| 1997 |
+
"loss": 1.5421,
|
| 1998 |
+
"step": 14200
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.8635055294652325,
|
| 2002 |
+
"grad_norm": 1.622070074081421,
|
| 2003 |
+
"learning_rate": 1.368236078288796e-06,
|
| 2004 |
+
"loss": 1.5283,
|
| 2005 |
+
"step": 14250
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.8665353734282685,
|
| 2009 |
+
"grad_norm": 1.5320727825164795,
|
| 2010 |
+
"learning_rate": 1.3379385566260681e-06,
|
| 2011 |
+
"loss": 1.5386,
|
| 2012 |
+
"step": 14300
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.8695652173913043,
|
| 2016 |
+
"grad_norm": 1.5714218616485596,
|
| 2017 |
+
"learning_rate": 1.30764103496334e-06,
|
| 2018 |
+
"loss": 1.558,
|
| 2019 |
+
"step": 14350
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.8725950613543403,
|
| 2023 |
+
"grad_norm": 1.5528987646102905,
|
| 2024 |
+
"learning_rate": 1.277343513300612e-06,
|
| 2025 |
+
"loss": 1.5058,
|
| 2026 |
+
"step": 14400
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.8756249053173761,
|
| 2030 |
+
"grad_norm": 1.4519600868225098,
|
| 2031 |
+
"learning_rate": 1.2470459916378842e-06,
|
| 2032 |
+
"loss": 1.535,
|
| 2033 |
+
"step": 14450
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.8786547492804121,
|
| 2037 |
+
"grad_norm": 1.9346317052841187,
|
| 2038 |
+
"learning_rate": 1.2173544204084107e-06,
|
| 2039 |
+
"loss": 1.5321,
|
| 2040 |
+
"step": 14500
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.881684593243448,
|
| 2044 |
+
"grad_norm": 1.4452751874923706,
|
| 2045 |
+
"learning_rate": 1.1870568987456827e-06,
|
| 2046 |
+
"loss": 1.5147,
|
| 2047 |
+
"step": 14550
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.8847144372064839,
|
| 2051 |
+
"grad_norm": 1.4155831336975098,
|
| 2052 |
+
"learning_rate": 1.1567593770829546e-06,
|
| 2053 |
+
"loss": 1.5439,
|
| 2054 |
+
"step": 14600
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.8877442811695198,
|
| 2058 |
+
"grad_norm": 1.3461506366729736,
|
| 2059 |
+
"learning_rate": 1.1264618554202268e-06,
|
| 2060 |
+
"loss": 1.517,
|
| 2061 |
+
"step": 14650
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.8907741251325557,
|
| 2065 |
+
"grad_norm": 1.4958879947662354,
|
| 2066 |
+
"learning_rate": 1.0961643337574987e-06,
|
| 2067 |
+
"loss": 1.5288,
|
| 2068 |
+
"step": 14700
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.8938039690955916,
|
| 2072 |
+
"grad_norm": 1.610373616218567,
|
| 2073 |
+
"learning_rate": 1.0658668120947706e-06,
|
| 2074 |
+
"loss": 1.5272,
|
| 2075 |
+
"step": 14750
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.8968338130586275,
|
| 2079 |
+
"grad_norm": 1.5633084774017334,
|
| 2080 |
+
"learning_rate": 1.0355692904320428e-06,
|
| 2081 |
+
"loss": 1.524,
|
| 2082 |
+
"step": 14800
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.8998636570216634,
|
| 2086 |
+
"grad_norm": 1.5080559253692627,
|
| 2087 |
+
"learning_rate": 1.0052717687693147e-06,
|
| 2088 |
+
"loss": 1.5302,
|
| 2089 |
+
"step": 14850
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.9028935009846993,
|
| 2093 |
+
"grad_norm": 1.5691548585891724,
|
| 2094 |
+
"learning_rate": 9.749742471065869e-07,
|
| 2095 |
+
"loss": 1.5573,
|
| 2096 |
+
"step": 14900
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.9059233449477352,
|
| 2100 |
+
"grad_norm": 1.44107985496521,
|
| 2101 |
+
"learning_rate": 9.446767254438588e-07,
|
| 2102 |
+
"loss": 1.5266,
|
| 2103 |
+
"step": 14950
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.9089531889107711,
|
| 2107 |
+
"grad_norm": 1.5320124626159668,
|
| 2108 |
+
"learning_rate": 9.143792037811309e-07,
|
| 2109 |
+
"loss": 1.5842,
|
| 2110 |
+
"step": 15000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.911983032873807,
|
| 2114 |
+
"grad_norm": 1.8644306659698486,
|
| 2115 |
+
"learning_rate": 8.840816821184028e-07,
|
| 2116 |
+
"loss": 1.5547,
|
| 2117 |
+
"step": 15050
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.9150128768368428,
|
| 2121 |
+
"grad_norm": 1.5941940546035767,
|
| 2122 |
+
"learning_rate": 8.537841604556748e-07,
|
| 2123 |
+
"loss": 1.5481,
|
| 2124 |
+
"step": 15100
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.9180427207998788,
|
| 2128 |
+
"grad_norm": 1.3055516481399536,
|
| 2129 |
+
"learning_rate": 8.234866387929468e-07,
|
| 2130 |
+
"loss": 1.5227,
|
| 2131 |
+
"step": 15150
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.9210725647629147,
|
| 2135 |
+
"grad_norm": 1.5031893253326416,
|
| 2136 |
+
"learning_rate": 7.931891171302187e-07,
|
| 2137 |
+
"loss": 1.5489,
|
| 2138 |
+
"step": 15200
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.9241024087259506,
|
| 2142 |
+
"grad_norm": 1.575303554534912,
|
| 2143 |
+
"learning_rate": 7.628915954674908e-07,
|
| 2144 |
+
"loss": 1.5486,
|
| 2145 |
+
"step": 15250
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.9271322526889865,
|
| 2149 |
+
"grad_norm": 1.5559568405151367,
|
| 2150 |
+
"learning_rate": 7.325940738047628e-07,
|
| 2151 |
+
"loss": 1.4914,
|
| 2152 |
+
"step": 15300
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.9301620966520224,
|
| 2156 |
+
"grad_norm": 1.3393162488937378,
|
| 2157 |
+
"learning_rate": 7.022965521420348e-07,
|
| 2158 |
+
"loss": 1.5474,
|
| 2159 |
+
"step": 15350
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.9331919406150583,
|
| 2163 |
+
"grad_norm": 1.404003381729126,
|
| 2164 |
+
"learning_rate": 6.719990304793069e-07,
|
| 2165 |
+
"loss": 1.5088,
|
| 2166 |
+
"step": 15400
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.9362217845780942,
|
| 2170 |
+
"grad_norm": 1.5367882251739502,
|
| 2171 |
+
"learning_rate": 6.417015088165788e-07,
|
| 2172 |
+
"loss": 1.5545,
|
| 2173 |
+
"step": 15450
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.9392516285411301,
|
| 2177 |
+
"grad_norm": 1.4406241178512573,
|
| 2178 |
+
"learning_rate": 6.114039871538509e-07,
|
| 2179 |
+
"loss": 1.527,
|
| 2180 |
+
"step": 15500
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.942281472504166,
|
| 2184 |
+
"grad_norm": 1.3530008792877197,
|
| 2185 |
+
"learning_rate": 5.811064654911229e-07,
|
| 2186 |
+
"loss": 1.5504,
|
| 2187 |
+
"step": 15550
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.9453113164672019,
|
| 2191 |
+
"grad_norm": 1.4533566236495972,
|
| 2192 |
+
"learning_rate": 5.508089438283949e-07,
|
| 2193 |
+
"loss": 1.518,
|
| 2194 |
+
"step": 15600
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.9483411604302379,
|
| 2198 |
+
"grad_norm": 1.5848952531814575,
|
| 2199 |
+
"learning_rate": 5.205114221656669e-07,
|
| 2200 |
+
"loss": 1.524,
|
| 2201 |
+
"step": 15650
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.9513710043932737,
|
| 2205 |
+
"grad_norm": 1.5302923917770386,
|
| 2206 |
+
"learning_rate": 4.902139005029388e-07,
|
| 2207 |
+
"loss": 1.5194,
|
| 2208 |
+
"step": 15700
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.9544008483563097,
|
| 2212 |
+
"grad_norm": 1.381064772605896,
|
| 2213 |
+
"learning_rate": 4.599163788402109e-07,
|
| 2214 |
+
"loss": 1.5345,
|
| 2215 |
+
"step": 15750
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.9574306923193455,
|
| 2219 |
+
"grad_norm": 1.5697840452194214,
|
| 2220 |
+
"learning_rate": 4.2961885717748293e-07,
|
| 2221 |
+
"loss": 1.5332,
|
| 2222 |
+
"step": 15800
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.9604605362823815,
|
| 2226 |
+
"grad_norm": 1.5636427402496338,
|
| 2227 |
+
"learning_rate": 3.99321335514755e-07,
|
| 2228 |
+
"loss": 1.5104,
|
| 2229 |
+
"step": 15850
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.9634903802454173,
|
| 2233 |
+
"grad_norm": 1.45536208152771,
|
| 2234 |
+
"learning_rate": 3.690238138520269e-07,
|
| 2235 |
+
"loss": 1.5505,
|
| 2236 |
+
"step": 15900
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.9665202242084533,
|
| 2240 |
+
"grad_norm": 1.49254310131073,
|
| 2241 |
+
"learning_rate": 3.387262921892989e-07,
|
| 2242 |
+
"loss": 1.512,
|
| 2243 |
+
"step": 15950
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.9695500681714891,
|
| 2247 |
+
"grad_norm": 1.603808045387268,
|
| 2248 |
+
"learning_rate": 3.0842877052657095e-07,
|
| 2249 |
+
"loss": 1.5309,
|
| 2250 |
+
"step": 16000
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.9725799121345251,
|
| 2254 |
+
"grad_norm": 1.5627918243408203,
|
| 2255 |
+
"learning_rate": 2.7813124886384294e-07,
|
| 2256 |
+
"loss": 1.5498,
|
| 2257 |
+
"step": 16050
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.975609756097561,
|
| 2261 |
+
"grad_norm": 1.4155369997024536,
|
| 2262 |
+
"learning_rate": 2.47833727201115e-07,
|
| 2263 |
+
"loss": 1.567,
|
| 2264 |
+
"step": 16100
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.9786396000605969,
|
| 2268 |
+
"grad_norm": 1.3787842988967896,
|
| 2269 |
+
"learning_rate": 2.1753620553838698e-07,
|
| 2270 |
+
"loss": 1.5511,
|
| 2271 |
+
"step": 16150
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.9816694440236328,
|
| 2275 |
+
"grad_norm": 1.3937129974365234,
|
| 2276 |
+
"learning_rate": 1.87238683875659e-07,
|
| 2277 |
+
"loss": 1.4971,
|
| 2278 |
+
"step": 16200
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.9846992879866687,
|
| 2282 |
+
"grad_norm": 1.588771104812622,
|
| 2283 |
+
"learning_rate": 1.5694116221293101e-07,
|
| 2284 |
+
"loss": 1.5384,
|
| 2285 |
+
"step": 16250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.9877291319497046,
|
| 2289 |
+
"grad_norm": 1.402185320854187,
|
| 2290 |
+
"learning_rate": 1.26643640550203e-07,
|
| 2291 |
+
"loss": 1.517,
|
| 2292 |
+
"step": 16300
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.9907589759127405,
|
| 2296 |
+
"grad_norm": 1.557084321975708,
|
| 2297 |
+
"learning_rate": 9.634611888747501e-08,
|
| 2298 |
+
"loss": 1.5527,
|
| 2299 |
+
"step": 16350
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.9937888198757764,
|
| 2303 |
+
"grad_norm": 1.4594271183013916,
|
| 2304 |
+
"learning_rate": 6.604859722474703e-08,
|
| 2305 |
+
"loss": 1.5216,
|
| 2306 |
+
"step": 16400
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.9968186638388123,
|
| 2310 |
+
"grad_norm": 1.467354655265808,
|
| 2311 |
+
"learning_rate": 3.5751075562019034e-08,
|
| 2312 |
+
"loss": 1.5684,
|
| 2313 |
+
"step": 16450
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.9998485078018482,
|
| 2317 |
+
"grad_norm": 1.326578140258789,
|
| 2318 |
+
"learning_rate": 5.453553899291039e-09,
|
| 2319 |
+
"loss": 1.5009,
|
| 2320 |
+
"step": 16500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.0,
|
| 2324 |
+
"eval_loss": 1.4647282361984253,
|
| 2325 |
+
"eval_runtime": 301.5458,
|
| 2326 |
+
"eval_samples_per_second": 48.646,
|
| 2327 |
+
"eval_steps_per_second": 6.082,
|
| 2328 |
+
"step": 16503
|
| 2329 |
+
}
|
| 2330 |
+
],
|
| 2331 |
+
"logging_steps": 50,
|
| 2332 |
+
"max_steps": 16503,
|
| 2333 |
+
"num_input_tokens_seen": 0,
|
| 2334 |
+
"num_train_epochs": 1,
|
| 2335 |
+
"save_steps": 500,
|
| 2336 |
+
"stateful_callbacks": {
|
| 2337 |
+
"TrainerControl": {
|
| 2338 |
+
"args": {
|
| 2339 |
+
"should_epoch_stop": false,
|
| 2340 |
+
"should_evaluate": false,
|
| 2341 |
+
"should_log": false,
|
| 2342 |
+
"should_save": true,
|
| 2343 |
+
"should_training_stop": true
|
| 2344 |
+
},
|
| 2345 |
+
"attributes": {}
|
| 2346 |
+
}
|
| 2347 |
+
},
|
| 2348 |
+
"total_flos": 3.4495512772608e+16,
|
| 2349 |
+
"train_batch_size": 2,
|
| 2350 |
+
"trial_name": null,
|
| 2351 |
+
"trial_params": null
|
| 2352 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13238729d22193b67545fd1c422a58623572ca7998d2129683f5cee09768dc72
|
| 3 |
+
size 5368
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|