Training in progress, step 21362
Browse files- checkpoint-21362/config.json +39 -0
- checkpoint-21362/merges.txt +0 -0
- checkpoint-21362/optimizer.pt +3 -0
- checkpoint-21362/pytorch_model.bin +3 -0
- checkpoint-21362/rng_state.pth +3 -0
- checkpoint-21362/scaler.pt +3 -0
- checkpoint-21362/scheduler.pt +3 -0
- checkpoint-21362/special_tokens_map.json +6 -0
- checkpoint-21362/tokenizer.json +0 -0
- checkpoint-21362/tokenizer_config.json +10 -0
- checkpoint-21362/trainer_state.json +3152 -0
- checkpoint-21362/training_args.bin +3 -0
- checkpoint-21362/vocab.json +0 -0
- config.json +39 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +10 -0
- training_args.bin +3 -0
- vocab.json +0 -0
checkpoint-21362/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMAndValueHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": true,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.23.0",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50257
|
| 39 |
+
}
|
checkpoint-21362/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-21362/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b3a848e6ef34f45ade9a6349d3c32e42484cfd019d41349cc3bbdb7fbc35845
|
| 3 |
+
size 995605189
|
checkpoint-21362/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7f39fa26de841da6fe89695b4819d6ad5a6ddff087f7d0d2d506360443757a3
|
| 3 |
+
size 510398013
|
checkpoint-21362/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e46d0d51c9d952e32c0fbef91aa7ae68815b1d0195d8150cc3570dadae1c908
|
| 3 |
+
size 15661
|
checkpoint-21362/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4fc1fe765eb686e92dd7db4c791258cf0dd99cefbc8c069ed52f2985d70a3d5
|
| 3 |
+
size 557
|
checkpoint-21362/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c40bd496e66e8ad05f65c1e3f67e7b5e44108dc272dad9d6b03d42dd9a10017
|
| 3 |
+
size 627
|
checkpoint-21362/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
checkpoint-21362/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-21362/tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<|endoftext|>",
|
| 4 |
+
"eos_token": "<|endoftext|>",
|
| 5 |
+
"model_max_length": 1024,
|
| 6 |
+
"name_or_path": "gpt2",
|
| 7 |
+
"special_tokens_map_file": null,
|
| 8 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 9 |
+
"unk_token": "<|endoftext|>"
|
| 10 |
+
}
|
checkpoint-21362/trainer_state.json
ADDED
|
@@ -0,0 +1,3152 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5,
|
| 5 |
+
"global_step": 21362,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 1.6355140186915887e-06,
|
| 13 |
+
"loss": 10.7543,
|
| 14 |
+
"theoretical_loss": 20.81281780154715,
|
| 15 |
+
"tokens_seen": 65536
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.0,
|
| 19 |
+
"learning_rate": 8.177570093457944e-05,
|
| 20 |
+
"loss": 8.6636,
|
| 21 |
+
"theoretical_loss": 8.563482664611069,
|
| 22 |
+
"tokens_seen": 3276800
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.0,
|
| 26 |
+
"learning_rate": 0.0001635514018691589,
|
| 27 |
+
"loss": 6.6432,
|
| 28 |
+
"theoretical_loss": 7.4777587180480305,
|
| 29 |
+
"tokens_seen": 6553600
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.0,
|
| 33 |
+
"learning_rate": 0.0002453271028037383,
|
| 34 |
+
"loss": 5.9475,
|
| 35 |
+
"theoretical_loss": 6.9337544888949,
|
| 36 |
+
"tokens_seen": 9830400
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.0,
|
| 40 |
+
"learning_rate": 0.0003271028037383178,
|
| 41 |
+
"loss": 5.5534,
|
| 42 |
+
"theoretical_loss": 6.583566228426414,
|
| 43 |
+
"tokens_seen": 13107200
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 0.01,
|
| 47 |
+
"learning_rate": 0.0004088785046728972,
|
| 48 |
+
"loss": 5.3664,
|
| 49 |
+
"theoretical_loss": 6.330713565116083,
|
| 50 |
+
"tokens_seen": 16384000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.01,
|
| 54 |
+
"learning_rate": 0.0004906542056074766,
|
| 55 |
+
"loss": 5.1899,
|
| 56 |
+
"theoretical_loss": 6.135529231940326,
|
| 57 |
+
"tokens_seen": 19660800
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.01,
|
| 61 |
+
"learning_rate": 0.0005724299065420561,
|
| 62 |
+
"loss": 5.0234,
|
| 63 |
+
"theoretical_loss": 5.978101583869607,
|
| 64 |
+
"tokens_seen": 22937600
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.01,
|
| 68 |
+
"learning_rate": 0.0006542056074766356,
|
| 69 |
+
"loss": 4.9249,
|
| 70 |
+
"theoretical_loss": 5.8471173262659235,
|
| 71 |
+
"tokens_seen": 26214400
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.01,
|
| 75 |
+
"learning_rate": 0.0006996358993758275,
|
| 76 |
+
"loss": 4.7711,
|
| 77 |
+
"theoretical_loss": 5.7355768158821245,
|
| 78 |
+
"tokens_seen": 29491200
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.01,
|
| 82 |
+
"learning_rate": 0.0006988083979572536,
|
| 83 |
+
"loss": 4.7076,
|
| 84 |
+
"theoretical_loss": 5.638870144071353,
|
| 85 |
+
"tokens_seen": 32768000
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.01,
|
| 89 |
+
"learning_rate": 0.0006979808965386797,
|
| 90 |
+
"loss": 4.5841,
|
| 91 |
+
"theoretical_loss": 5.553812381844907,
|
| 92 |
+
"tokens_seen": 36044800
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.01,
|
| 96 |
+
"learning_rate": 0.000697153395120106,
|
| 97 |
+
"loss": 4.5051,
|
| 98 |
+
"theoretical_loss": 5.478118080556438,
|
| 99 |
+
"tokens_seen": 39321600
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.02,
|
| 103 |
+
"learning_rate": 0.0006963258937015321,
|
| 104 |
+
"loss": 4.4155,
|
| 105 |
+
"theoretical_loss": 5.410095959579362,
|
| 106 |
+
"tokens_seen": 42598400
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.02,
|
| 110 |
+
"learning_rate": 0.0006954983922829582,
|
| 111 |
+
"loss": 4.3751,
|
| 112 |
+
"theoretical_loss": 5.348462083735834,
|
| 113 |
+
"tokens_seen": 45875200
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.02,
|
| 117 |
+
"learning_rate": 0.0006946708908643843,
|
| 118 |
+
"loss": 4.2199,
|
| 119 |
+
"theoretical_loss": 5.292220566937567,
|
| 120 |
+
"tokens_seen": 49152000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.02,
|
| 124 |
+
"learning_rate": 0.0006938433894458105,
|
| 125 |
+
"loss": 4.189,
|
| 126 |
+
"theoretical_loss": 5.240584625769978,
|
| 127 |
+
"tokens_seen": 52428800
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 0.02,
|
| 131 |
+
"learning_rate": 0.0006930158880272367,
|
| 132 |
+
"loss": 4.1234,
|
| 133 |
+
"theoretical_loss": 5.192922724525789,
|
| 134 |
+
"tokens_seen": 55705600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.02,
|
| 138 |
+
"learning_rate": 0.0006921883866086628,
|
| 139 |
+
"loss": 4.051,
|
| 140 |
+
"theoretical_loss": 5.1487208633564405,
|
| 141 |
+
"tokens_seen": 58982400
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.02,
|
| 145 |
+
"learning_rate": 0.0006913608851900889,
|
| 146 |
+
"loss": 3.9171,
|
| 147 |
+
"theoretical_loss": 5.107555562405102,
|
| 148 |
+
"tokens_seen": 62259200
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.02,
|
| 152 |
+
"learning_rate": 0.000690533383771515,
|
| 153 |
+
"loss": 3.8106,
|
| 154 |
+
"theoretical_loss": 5.069074117143246,
|
| 155 |
+
"tokens_seen": 65536000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.02,
|
| 159 |
+
"learning_rate": 0.0006897058823529412,
|
| 160 |
+
"loss": 3.7893,
|
| 161 |
+
"theoretical_loss": 5.032979909838007,
|
| 162 |
+
"tokens_seen": 68812800
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.03,
|
| 166 |
+
"learning_rate": 0.0006888783809343674,
|
| 167 |
+
"loss": 3.7301,
|
| 168 |
+
"theoretical_loss": 4.999021308224664,
|
| 169 |
+
"tokens_seen": 72089600
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.03,
|
| 173 |
+
"learning_rate": 0.0006880508795157935,
|
| 174 |
+
"loss": 3.6744,
|
| 175 |
+
"theoretical_loss": 4.966983155351962,
|
| 176 |
+
"tokens_seen": 75366400
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.03,
|
| 180 |
+
"learning_rate": 0.0006872233780972196,
|
| 181 |
+
"loss": 3.5873,
|
| 182 |
+
"theoretical_loss": 4.9366801616251355,
|
| 183 |
+
"tokens_seen": 78643200
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.03,
|
| 187 |
+
"learning_rate": 0.0006863958766786457,
|
| 188 |
+
"loss": 3.5755,
|
| 189 |
+
"theoretical_loss": 4.907951713830082,
|
| 190 |
+
"tokens_seen": 81920000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.03,
|
| 194 |
+
"learning_rate": 0.0006855683752600718,
|
| 195 |
+
"loss": 3.5132,
|
| 196 |
+
"theoretical_loss": 4.880657753812926,
|
| 197 |
+
"tokens_seen": 85196800
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 0.03,
|
| 201 |
+
"learning_rate": 0.000684740873841498,
|
| 202 |
+
"loss": 3.5089,
|
| 203 |
+
"theoretical_loss": 4.854675474481779,
|
| 204 |
+
"tokens_seen": 88473600
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.03,
|
| 208 |
+
"learning_rate": 0.0006839133724229242,
|
| 209 |
+
"loss": 3.4237,
|
| 210 |
+
"theoretical_loss": 4.8298966473088125,
|
| 211 |
+
"tokens_seen": 91750400
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.03,
|
| 215 |
+
"learning_rate": 0.0006830858710043503,
|
| 216 |
+
"loss": 3.4596,
|
| 217 |
+
"theoretical_loss": 4.8062254427779205,
|
| 218 |
+
"tokens_seen": 95027200
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.04,
|
| 222 |
+
"learning_rate": 0.0006822583695857764,
|
| 223 |
+
"loss": 3.4046,
|
| 224 |
+
"theoretical_loss": 4.783576639276257,
|
| 225 |
+
"tokens_seen": 98304000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.04,
|
| 229 |
+
"learning_rate": 0.0006814308681672025,
|
| 230 |
+
"loss": 3.4409,
|
| 231 |
+
"theoretical_loss": 4.761874140772408,
|
| 232 |
+
"tokens_seen": 101580800
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.04,
|
| 236 |
+
"learning_rate": 0.0006806033667486286,
|
| 237 |
+
"loss": 3.4431,
|
| 238 |
+
"theoretical_loss": 4.741049741962473,
|
| 239 |
+
"tokens_seen": 104857600
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.04,
|
| 243 |
+
"learning_rate": 0.0006797758653300548,
|
| 244 |
+
"loss": 3.4137,
|
| 245 |
+
"theoretical_loss": 4.721042093249051,
|
| 246 |
+
"tokens_seen": 108134400
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.04,
|
| 250 |
+
"learning_rate": 0.000678948363911481,
|
| 251 |
+
"loss": 3.3936,
|
| 252 |
+
"theoretical_loss": 4.701795828231866,
|
| 253 |
+
"tokens_seen": 111411200
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.04,
|
| 257 |
+
"learning_rate": 0.0006781208624929071,
|
| 258 |
+
"loss": 3.3917,
|
| 259 |
+
"theoretical_loss": 4.68326082423593,
|
| 260 |
+
"tokens_seen": 114688000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.04,
|
| 264 |
+
"learning_rate": 0.0006772933610743332,
|
| 265 |
+
"loss": 3.3676,
|
| 266 |
+
"theoretical_loss": 4.665391572426282,
|
| 267 |
+
"tokens_seen": 117964800
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.04,
|
| 271 |
+
"learning_rate": 0.0006764658596557593,
|
| 272 |
+
"loss": 3.3325,
|
| 273 |
+
"theoretical_loss": 4.648146638719739,
|
| 274 |
+
"tokens_seen": 121241600
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.04,
|
| 278 |
+
"learning_rate": 0.0006756383582371856,
|
| 279 |
+
"loss": 3.345,
|
| 280 |
+
"theoretical_loss": 4.631488200339643,
|
| 281 |
+
"tokens_seen": 124518400
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.05,
|
| 285 |
+
"learning_rate": 0.0006748108568186117,
|
| 286 |
+
"loss": 3.371,
|
| 287 |
+
"theoretical_loss": 4.615381645715717,
|
| 288 |
+
"tokens_seen": 127795200
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.05,
|
| 292 |
+
"learning_rate": 0.0006739833554000378,
|
| 293 |
+
"loss": 3.2953,
|
| 294 |
+
"theoretical_loss": 4.599795227690505,
|
| 295 |
+
"tokens_seen": 131072000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.05,
|
| 299 |
+
"learning_rate": 0.000673155853981464,
|
| 300 |
+
"loss": 3.3211,
|
| 301 |
+
"theoretical_loss": 4.584699761792674,
|
| 302 |
+
"tokens_seen": 134348800
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.05,
|
| 306 |
+
"learning_rate": 0.0006723283525628902,
|
| 307 |
+
"loss": 3.2745,
|
| 308 |
+
"theoretical_loss": 4.570068362778516,
|
| 309 |
+
"tokens_seen": 137625600
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.05,
|
| 313 |
+
"learning_rate": 0.0006715008511443163,
|
| 314 |
+
"loss": 3.289,
|
| 315 |
+
"theoretical_loss": 4.555876213804037,
|
| 316 |
+
"tokens_seen": 140902400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.05,
|
| 320 |
+
"learning_rate": 0.0006706733497257424,
|
| 321 |
+
"loss": 3.2753,
|
| 322 |
+
"theoretical_loss": 4.542100363530799,
|
| 323 |
+
"tokens_seen": 144179200
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 0.05,
|
| 327 |
+
"learning_rate": 0.0006698458483071685,
|
| 328 |
+
"loss": 3.2546,
|
| 329 |
+
"theoretical_loss": 4.528719547234816,
|
| 330 |
+
"tokens_seen": 147456000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.05,
|
| 334 |
+
"learning_rate": 0.0006690183468885946,
|
| 335 |
+
"loss": 3.2145,
|
| 336 |
+
"theoretical_loss": 4.515714028614996,
|
| 337 |
+
"tokens_seen": 150732800
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.06,
|
| 341 |
+
"learning_rate": 0.0006681908454700209,
|
| 342 |
+
"loss": 3.2252,
|
| 343 |
+
"theoretical_loss": 4.503065459513339,
|
| 344 |
+
"tokens_seen": 154009600
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.06,
|
| 348 |
+
"learning_rate": 0.000667363344051447,
|
| 349 |
+
"loss": 3.2379,
|
| 350 |
+
"theoretical_loss": 4.4907567551852665,
|
| 351 |
+
"tokens_seen": 157286400
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.06,
|
| 355 |
+
"learning_rate": 0.0006665358426328731,
|
| 356 |
+
"loss": 3.2175,
|
| 357 |
+
"theoretical_loss": 4.478771983111967,
|
| 358 |
+
"tokens_seen": 160563200
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.06,
|
| 362 |
+
"objective/train/avg_token_score": 0.027871694415807724,
|
| 363 |
+
"objective/train/avg_weight": 0.9700481295585632,
|
| 364 |
+
"objective/train/docs_used": 104000,
|
| 365 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 366 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 367 |
+
"objective/train/original_loss": 3.27850341796875,
|
| 368 |
+
"objective/train/std_weight": 0.08482376486063004,
|
| 369 |
+
"objective/train/theoretical_loss": 4.467096263641219,
|
| 370 |
+
"objective/train/tokens_used": 184300000,
|
| 371 |
+
"theoretical_loss": 4.467096263641219,
|
| 372 |
+
"tokens_seen": 163840000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 0.06,
|
| 376 |
+
"learning_rate": 0.0006657083412142992,
|
| 377 |
+
"loss": 3.2065,
|
| 378 |
+
"theoretical_loss": 4.467096263641219,
|
| 379 |
+
"tokens_seen": 163840000
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 0.06,
|
| 383 |
+
"learning_rate": 0.0006648808397957253,
|
| 384 |
+
"loss": 3.2165,
|
| 385 |
+
"theoretical_loss": 4.455715680989545,
|
| 386 |
+
"tokens_seen": 167116800
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.06,
|
| 390 |
+
"learning_rate": 0.0006640533383771514,
|
| 391 |
+
"loss": 3.168,
|
| 392 |
+
"theoretical_loss": 4.44461720334543,
|
| 393 |
+
"tokens_seen": 170393600
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.06,
|
| 397 |
+
"learning_rate": 0.0006632258369585777,
|
| 398 |
+
"loss": 3.118,
|
| 399 |
+
"theoretical_loss": 4.433788610987646,
|
| 400 |
+
"tokens_seen": 173670400
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.06,
|
| 404 |
+
"learning_rate": 0.0006623983355400038,
|
| 405 |
+
"loss": 3.1702,
|
| 406 |
+
"theoretical_loss": 4.42321843148016,
|
| 407 |
+
"tokens_seen": 176947200
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 0.06,
|
| 411 |
+
"learning_rate": 0.0006615708341214299,
|
| 412 |
+
"loss": 3.1235,
|
| 413 |
+
"theoretical_loss": 4.412895881130142,
|
| 414 |
+
"tokens_seen": 180224000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 0.07,
|
| 418 |
+
"learning_rate": 0.000660743332702856,
|
| 419 |
+
"loss": 3.1592,
|
| 420 |
+
"theoretical_loss": 4.4028108120020795,
|
| 421 |
+
"tokens_seen": 183500800
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 0.07,
|
| 425 |
+
"learning_rate": 0.0006599158312842821,
|
| 426 |
+
"loss": 3.1533,
|
| 427 |
+
"theoretical_loss": 4.392953663871862,
|
| 428 |
+
"tokens_seen": 186777600
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.07,
|
| 432 |
+
"learning_rate": 0.0006590883298657083,
|
| 433 |
+
"loss": 3.1398,
|
| 434 |
+
"theoretical_loss": 4.383315420582533,
|
| 435 |
+
"tokens_seen": 190054400
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.07,
|
| 439 |
+
"learning_rate": 0.0006582608284471345,
|
| 440 |
+
"loss": 3.1768,
|
| 441 |
+
"theoretical_loss": 4.373887570330275,
|
| 442 |
+
"tokens_seen": 193331200
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.07,
|
| 446 |
+
"learning_rate": 0.0006574333270285606,
|
| 447 |
+
"loss": 3.136,
|
| 448 |
+
"theoretical_loss": 4.364662069466704,
|
| 449 |
+
"tokens_seen": 196608000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.07,
|
| 453 |
+
"learning_rate": 0.0006566058256099867,
|
| 454 |
+
"loss": 3.134,
|
| 455 |
+
"theoretical_loss": 4.355631309453283,
|
| 456 |
+
"tokens_seen": 199884800
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.07,
|
| 460 |
+
"learning_rate": 0.0006557783241914128,
|
| 461 |
+
"loss": 3.0886,
|
| 462 |
+
"theoretical_loss": 4.346788086646671,
|
| 463 |
+
"tokens_seen": 203161600
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.07,
|
| 467 |
+
"learning_rate": 0.0006549508227728391,
|
| 468 |
+
"loss": 3.1173,
|
| 469 |
+
"theoretical_loss": 4.33812557463116,
|
| 470 |
+
"tokens_seen": 206438400
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.07,
|
| 474 |
+
"learning_rate": 0.0006541233213542652,
|
| 475 |
+
"loss": 3.1278,
|
| 476 |
+
"theoretical_loss": 4.329637298846812,
|
| 477 |
+
"tokens_seen": 209715200
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.08,
|
| 481 |
+
"learning_rate": 0.0006532958199356913,
|
| 482 |
+
"loss": 3.0931,
|
| 483 |
+
"theoretical_loss": 4.321317113290252,
|
| 484 |
+
"tokens_seen": 212992000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.08,
|
| 488 |
+
"learning_rate": 0.0006524683185171174,
|
| 489 |
+
"loss": 3.0638,
|
| 490 |
+
"theoretical_loss": 4.3131591790897925,
|
| 491 |
+
"tokens_seen": 216268800
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.08,
|
| 495 |
+
"learning_rate": 0.0006516408170985437,
|
| 496 |
+
"loss": 3.1152,
|
| 497 |
+
"theoretical_loss": 4.305157944778228,
|
| 498 |
+
"tokens_seen": 219545600
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.08,
|
| 502 |
+
"learning_rate": 0.0006508133156799698,
|
| 503 |
+
"loss": 3.1342,
|
| 504 |
+
"theoretical_loss": 4.297308128105687,
|
| 505 |
+
"tokens_seen": 222822400
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.08,
|
| 509 |
+
"learning_rate": 0.0006499858142613959,
|
| 510 |
+
"loss": 3.1163,
|
| 511 |
+
"theoretical_loss": 4.2896046992515995,
|
| 512 |
+
"tokens_seen": 226099200
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.08,
|
| 516 |
+
"learning_rate": 0.000649158312842822,
|
| 517 |
+
"loss": 3.152,
|
| 518 |
+
"theoretical_loss": 4.282042865309616,
|
| 519 |
+
"tokens_seen": 229376000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.08,
|
| 523 |
+
"learning_rate": 0.0006483308114242481,
|
| 524 |
+
"loss": 3.1198,
|
| 525 |
+
"theoretical_loss": 4.274618055932298,
|
| 526 |
+
"tokens_seen": 232652800
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.08,
|
| 530 |
+
"learning_rate": 0.0006475033100056744,
|
| 531 |
+
"loss": 3.0873,
|
| 532 |
+
"theoretical_loss": 4.267325910033897,
|
| 533 |
+
"tokens_seen": 235929600
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.09,
|
| 537 |
+
"learning_rate": 0.0006466758085871005,
|
| 538 |
+
"loss": 3.0382,
|
| 539 |
+
"theoretical_loss": 4.260162263459744,
|
| 540 |
+
"tokens_seen": 239206400
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.09,
|
| 544 |
+
"learning_rate": 0.0006458483071685266,
|
| 545 |
+
"loss": 3.0672,
|
| 546 |
+
"theoretical_loss": 4.253123137539814,
|
| 547 |
+
"tokens_seen": 242483200
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.09,
|
| 551 |
+
"learning_rate": 0.0006450208057499527,
|
| 552 |
+
"loss": 3.0837,
|
| 553 |
+
"theoretical_loss": 4.246204728452055,
|
| 554 |
+
"tokens_seen": 245760000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.09,
|
| 558 |
+
"learning_rate": 0.0006441933043313788,
|
| 559 |
+
"loss": 3.1016,
|
| 560 |
+
"theoretical_loss": 4.239403397328261,
|
| 561 |
+
"tokens_seen": 249036800
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.09,
|
| 565 |
+
"learning_rate": 0.0006433658029128049,
|
| 566 |
+
"loss": 3.0698,
|
| 567 |
+
"theoretical_loss": 4.232715661041632,
|
| 568 |
+
"tokens_seen": 252313600
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 0.09,
|
| 572 |
+
"learning_rate": 0.0006425383014942312,
|
| 573 |
+
"loss": 3.0794,
|
| 574 |
+
"theoretical_loss": 4.226138183620867,
|
| 575 |
+
"tokens_seen": 255590400
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.09,
|
| 579 |
+
"learning_rate": 0.0006417108000756573,
|
| 580 |
+
"loss": 3.0643,
|
| 581 |
+
"theoretical_loss": 4.219667768240775,
|
| 582 |
+
"tokens_seen": 258867200
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 0.09,
|
| 586 |
+
"learning_rate": 0.0006408832986570834,
|
| 587 |
+
"loss": 3.0542,
|
| 588 |
+
"theoretical_loss": 4.213301349743924,
|
| 589 |
+
"tokens_seen": 262144000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 0.09,
|
| 593 |
+
"learning_rate": 0.0006400557972385095,
|
| 594 |
+
"loss": 2.9911,
|
| 595 |
+
"theoretical_loss": 4.20703598765197,
|
| 596 |
+
"tokens_seen": 265420800
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.1,
|
| 600 |
+
"learning_rate": 0.0006392282958199356,
|
| 601 |
+
"loss": 2.9473,
|
| 602 |
+
"theoretical_loss": 4.2008688596290025,
|
| 603 |
+
"tokens_seen": 268697600
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.1,
|
| 607 |
+
"learning_rate": 0.0006384007944013618,
|
| 608 |
+
"loss": 2.9018,
|
| 609 |
+
"theoretical_loss": 4.194797255362549,
|
| 610 |
+
"tokens_seen": 271974400
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.1,
|
| 614 |
+
"learning_rate": 0.000637573292982788,
|
| 615 |
+
"loss": 2.9774,
|
| 616 |
+
"theoretical_loss": 4.188818570830883,
|
| 617 |
+
"tokens_seen": 275251200
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.1,
|
| 621 |
+
"learning_rate": 0.0006367457915642141,
|
| 622 |
+
"loss": 2.9891,
|
| 623 |
+
"theoretical_loss": 4.182930302927963,
|
| 624 |
+
"tokens_seen": 278528000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.1,
|
| 628 |
+
"learning_rate": 0.0006359182901456402,
|
| 629 |
+
"loss": 2.952,
|
| 630 |
+
"theoretical_loss": 4.17713004441978,
|
| 631 |
+
"tokens_seen": 281804800
|
| 632 |
+
},
|
| 633 |
+
{
|
| 634 |
+
"epoch": 0.1,
|
| 635 |
+
"learning_rate": 0.0006350907887270663,
|
| 636 |
+
"loss": 2.9969,
|
| 637 |
+
"theoretical_loss": 4.1714154792080915,
|
| 638 |
+
"tokens_seen": 285081600
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.1,
|
| 642 |
+
"learning_rate": 0.0006342632873084925,
|
| 643 |
+
"loss": 3.0117,
|
| 644 |
+
"theoretical_loss": 4.165784377879517,
|
| 645 |
+
"tokens_seen": 288358400
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 0.1,
|
| 649 |
+
"learning_rate": 0.0006334357858899187,
|
| 650 |
+
"loss": 2.9764,
|
| 651 |
+
"theoretical_loss": 4.160234593519768,
|
| 652 |
+
"tokens_seen": 291635200
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 0.11,
|
| 656 |
+
"learning_rate": 0.0006326082844713448,
|
| 657 |
+
"loss": 2.997,
|
| 658 |
+
"theoretical_loss": 4.15476405777444,
|
| 659 |
+
"tokens_seen": 294912000
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 0.11,
|
| 663 |
+
"learning_rate": 0.0006317807830527709,
|
| 664 |
+
"loss": 2.9709,
|
| 665 |
+
"theoretical_loss": 4.149370777139286,
|
| 666 |
+
"tokens_seen": 298188800
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 0.11,
|
| 670 |
+
"learning_rate": 0.0006309532816341972,
|
| 671 |
+
"loss": 2.968,
|
| 672 |
+
"theoretical_loss": 4.144052829464249,
|
| 673 |
+
"tokens_seen": 301465600
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.11,
|
| 677 |
+
"learning_rate": 0.0006301257802156233,
|
| 678 |
+
"loss": 2.9237,
|
| 679 |
+
"theoretical_loss": 4.138808360656742,
|
| 680 |
+
"tokens_seen": 304742400
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.11,
|
| 684 |
+
"learning_rate": 0.0006292982787970494,
|
| 685 |
+
"loss": 2.9387,
|
| 686 |
+
"theoretical_loss": 4.133635581570836,
|
| 687 |
+
"tokens_seen": 308019200
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.11,
|
| 691 |
+
"learning_rate": 0.0006284707773784755,
|
| 692 |
+
"loss": 2.9275,
|
| 693 |
+
"theoretical_loss": 4.128532765070004,
|
| 694 |
+
"tokens_seen": 311296000
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 0.11,
|
| 698 |
+
"learning_rate": 0.0006276432759599016,
|
| 699 |
+
"loss": 2.9465,
|
| 700 |
+
"theoretical_loss": 4.123498243252032,
|
| 701 |
+
"tokens_seen": 314572800
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 0.11,
|
| 705 |
+
"learning_rate": 0.0006268157745413279,
|
| 706 |
+
"loss": 2.9295,
|
| 707 |
+
"theoretical_loss": 4.118530404825556,
|
| 708 |
+
"tokens_seen": 317849600
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.11,
|
| 712 |
+
"learning_rate": 0.000625988273122754,
|
| 713 |
+
"loss": 2.9456,
|
| 714 |
+
"theoretical_loss": 4.113627692628464,
|
| 715 |
+
"tokens_seen": 321126400
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 0.12,
|
| 719 |
+
"learning_rate": 0.0006251607717041801,
|
| 720 |
+
"loss": 2.949,
|
| 721 |
+
"theoretical_loss": 4.108788601279149,
|
| 722 |
+
"tokens_seen": 324403200
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"debugging/Self-BLEU-5": 0.5365128506817183,
|
| 726 |
+
"debugging/distinct-1-grams": 0.7612814402327299,
|
| 727 |
+
"debugging/distinct-2-grams": 0.9694583753853511,
|
| 728 |
+
"debugging/entropy-1-grams": 6.003629944255698,
|
| 729 |
+
"debugging/entropy-2-grams": 7.054987089269872,
|
| 730 |
+
"debugging/length": 495.25,
|
| 731 |
+
"debugging/num_segments": 16,
|
| 732 |
+
"epoch": 0.12,
|
| 733 |
+
"objective/train/avg_token_score": 0.04385810345411301,
|
| 734 |
+
"objective/train/avg_weight": 0.9528664350509644,
|
| 735 |
+
"objective/train/docs_used": 197327,
|
| 736 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 737 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 738 |
+
"objective/train/original_loss": 3.1396584510803223,
|
| 739 |
+
"objective/train/std_weight": 0.1686321198940277,
|
| 740 |
+
"objective/train/theoretical_loss": 4.10401167495222,
|
| 741 |
+
"objective/train/tokens_used": 348140000,
|
| 742 |
+
"theoretical_loss": 4.10401167495222,
|
| 743 |
+
"tokens_seen": 327680000
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.12,
|
| 747 |
+
"learning_rate": 0.0006243332702856062,
|
| 748 |
+
"loss": 2.9772,
|
| 749 |
+
"theoretical_loss": 4.10401167495222,
|
| 750 |
+
"tokens_seen": 327680000
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.12,
|
| 754 |
+
"learning_rate": 0.0006235057688670323,
|
| 755 |
+
"loss": 2.9764,
|
| 756 |
+
"theoretical_loss": 4.099295505270921,
|
| 757 |
+
"tokens_seen": 330956800
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.12,
|
| 761 |
+
"learning_rate": 0.0006226782674484584,
|
| 762 |
+
"loss": 2.9224,
|
| 763 |
+
"theoretical_loss": 4.094638729309031,
|
| 764 |
+
"tokens_seen": 334233600
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.12,
|
| 768 |
+
"learning_rate": 0.0006218507660298847,
|
| 769 |
+
"loss": 2.9348,
|
| 770 |
+
"theoretical_loss": 4.090040027695556,
|
| 771 |
+
"tokens_seen": 337510400
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.12,
|
| 775 |
+
"learning_rate": 0.0006210232646113108,
|
| 776 |
+
"loss": 2.8915,
|
| 777 |
+
"theoretical_loss": 4.085498122815992,
|
| 778 |
+
"tokens_seen": 340787200
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.12,
|
| 782 |
+
"learning_rate": 0.0006201957631927369,
|
| 783 |
+
"loss": 2.9147,
|
| 784 |
+
"theoretical_loss": 4.081011777104333,
|
| 785 |
+
"tokens_seen": 344064000
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.12,
|
| 789 |
+
"learning_rate": 0.000619368261774163,
|
| 790 |
+
"loss": 2.9209,
|
| 791 |
+
"theoretical_loss": 4.076579791420469,
|
| 792 |
+
"tokens_seen": 347340800
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.13,
|
| 796 |
+
"learning_rate": 0.0006185407603555891,
|
| 797 |
+
"loss": 2.9303,
|
| 798 |
+
"theoretical_loss": 4.0722010035079155,
|
| 799 |
+
"tokens_seen": 350617600
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.13,
|
| 803 |
+
"learning_rate": 0.0006177132589370153,
|
| 804 |
+
"loss": 2.9261,
|
| 805 |
+
"theoretical_loss": 4.067874286527197,
|
| 806 |
+
"tokens_seen": 353894400
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.13,
|
| 810 |
+
"learning_rate": 0.0006168857575184414,
|
| 811 |
+
"loss": 2.924,
|
| 812 |
+
"theoretical_loss": 4.063598547660519,
|
| 813 |
+
"tokens_seen": 357171200
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.13,
|
| 817 |
+
"learning_rate": 0.0006160582560998676,
|
| 818 |
+
"loss": 2.9546,
|
| 819 |
+
"theoretical_loss": 4.05937272678363,
|
| 820 |
+
"tokens_seen": 360448000
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.13,
|
| 824 |
+
"learning_rate": 0.0006152307546812937,
|
| 825 |
+
"loss": 2.9394,
|
| 826 |
+
"theoretical_loss": 4.055195795201069,
|
| 827 |
+
"tokens_seen": 363724800
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.13,
|
| 831 |
+
"learning_rate": 0.0006144032532627198,
|
| 832 |
+
"loss": 2.8913,
|
| 833 |
+
"theoretical_loss": 4.051066754441235,
|
| 834 |
+
"tokens_seen": 367001600
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.13,
|
| 838 |
+
"learning_rate": 0.000613575751844146,
|
| 839 |
+
"loss": 2.8973,
|
| 840 |
+
"theoretical_loss": 4.04698463510794,
|
| 841 |
+
"tokens_seen": 370278400
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.13,
|
| 845 |
+
"learning_rate": 0.0006127482504255721,
|
| 846 |
+
"loss": 2.901,
|
| 847 |
+
"theoretical_loss": 4.042948495785312,
|
| 848 |
+
"tokens_seen": 373555200
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.13,
|
| 852 |
+
"learning_rate": 0.0006119207490069983,
|
| 853 |
+
"loss": 2.8647,
|
| 854 |
+
"theoretical_loss": 4.038957421993153,
|
| 855 |
+
"tokens_seen": 376832000
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.14,
|
| 859 |
+
"learning_rate": 0.0006110932475884244,
|
| 860 |
+
"loss": 2.8707,
|
| 861 |
+
"theoretical_loss": 4.035010525189982,
|
| 862 |
+
"tokens_seen": 380108800
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.14,
|
| 866 |
+
"learning_rate": 0.0006102657461698505,
|
| 867 |
+
"loss": 2.904,
|
| 868 |
+
"theoretical_loss": 4.031106941821218,
|
| 869 |
+
"tokens_seen": 383385600
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.14,
|
| 873 |
+
"learning_rate": 0.0006094382447512768,
|
| 874 |
+
"loss": 2.8677,
|
| 875 |
+
"theoretical_loss": 4.027245832410079,
|
| 876 |
+
"tokens_seen": 386662400
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.14,
|
| 880 |
+
"learning_rate": 0.0006086107433327029,
|
| 881 |
+
"loss": 2.8421,
|
| 882 |
+
"theoretical_loss": 4.023426380688943,
|
| 883 |
+
"tokens_seen": 389939200
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.14,
|
| 887 |
+
"learning_rate": 0.000607783241914129,
|
| 888 |
+
"loss": 2.8849,
|
| 889 |
+
"theoretical_loss": 4.019647792769048,
|
| 890 |
+
"tokens_seen": 393216000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.14,
|
| 894 |
+
"learning_rate": 0.0006069557404955551,
|
| 895 |
+
"loss": 2.9447,
|
| 896 |
+
"theoretical_loss": 4.015909296346521,
|
| 897 |
+
"tokens_seen": 396492800
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.14,
|
| 901 |
+
"learning_rate": 0.0006061447891053528,
|
| 902 |
+
"loss": 2.9014,
|
| 903 |
+
"theoretical_loss": 4.012210139942894,
|
| 904 |
+
"tokens_seen": 399769600
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.14,
|
| 908 |
+
"learning_rate": 0.0006053172876867789,
|
| 909 |
+
"loss": 2.9463,
|
| 910 |
+
"theoretical_loss": 4.008549592178291,
|
| 911 |
+
"tokens_seen": 403046400
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.15,
|
| 915 |
+
"learning_rate": 0.000604489786268205,
|
| 916 |
+
"loss": 2.8978,
|
| 917 |
+
"theoretical_loss": 4.004926941075674,
|
| 918 |
+
"tokens_seen": 406323200
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.15,
|
| 922 |
+
"learning_rate": 0.0006036622848496311,
|
| 923 |
+
"loss": 2.9521,
|
| 924 |
+
"theoretical_loss": 4.001341493394558,
|
| 925 |
+
"tokens_seen": 409600000
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.15,
|
| 929 |
+
"learning_rate": 0.0006028347834310573,
|
| 930 |
+
"loss": 2.9652,
|
| 931 |
+
"theoretical_loss": 3.997792573992726,
|
| 932 |
+
"tokens_seen": 412876800
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.15,
|
| 936 |
+
"learning_rate": 0.0006020072820124835,
|
| 937 |
+
"loss": 2.9098,
|
| 938 |
+
"theoretical_loss": 3.994279525214554,
|
| 939 |
+
"tokens_seen": 416153600
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.15,
|
| 943 |
+
"learning_rate": 0.0006011797805939096,
|
| 944 |
+
"loss": 2.9572,
|
| 945 |
+
"theoretical_loss": 3.990801706304647,
|
| 946 |
+
"tokens_seen": 419430400
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.15,
|
| 950 |
+
"learning_rate": 0.0006003522791753358,
|
| 951 |
+
"loss": 2.9129,
|
| 952 |
+
"theoretical_loss": 3.987358492845532,
|
| 953 |
+
"tokens_seen": 422707200
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.15,
|
| 957 |
+
"learning_rate": 0.0005995247777567619,
|
| 958 |
+
"loss": 2.9439,
|
| 959 |
+
"theoretical_loss": 3.9839492762182647,
|
| 960 |
+
"tokens_seen": 425984000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.15,
|
| 964 |
+
"learning_rate": 0.000598697276338188,
|
| 965 |
+
"loss": 2.9134,
|
| 966 |
+
"theoretical_loss": 3.9805734630848306,
|
| 967 |
+
"tokens_seen": 429260800
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.15,
|
| 971 |
+
"learning_rate": 0.0005978697749196142,
|
| 972 |
+
"loss": 2.8965,
|
| 973 |
+
"theoretical_loss": 3.9772304748913054,
|
| 974 |
+
"tokens_seen": 432537600
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.16,
|
| 978 |
+
"learning_rate": 0.0005970422735010403,
|
| 979 |
+
"loss": 2.8898,
|
| 980 |
+
"theoretical_loss": 3.973919747390801,
|
| 981 |
+
"tokens_seen": 435814400
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.16,
|
| 985 |
+
"learning_rate": 0.0005962147720824664,
|
| 986 |
+
"loss": 2.9002,
|
| 987 |
+
"theoretical_loss": 3.9706407301852487,
|
| 988 |
+
"tokens_seen": 439091200
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.16,
|
| 992 |
+
"learning_rate": 0.0005953872706638926,
|
| 993 |
+
"loss": 2.8783,
|
| 994 |
+
"theoretical_loss": 3.9673928862851655,
|
| 995 |
+
"tokens_seen": 442368000
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.16,
|
| 999 |
+
"learning_rate": 0.0005945597692453187,
|
| 1000 |
+
"loss": 2.8682,
|
| 1001 |
+
"theoretical_loss": 3.9641756916865463,
|
| 1002 |
+
"tokens_seen": 445644800
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.16,
|
| 1006 |
+
"learning_rate": 0.0005937322678267449,
|
| 1007 |
+
"loss": 2.8783,
|
| 1008 |
+
"theoretical_loss": 3.960988634964113,
|
| 1009 |
+
"tokens_seen": 448921600
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.16,
|
| 1013 |
+
"learning_rate": 0.000592904766408171,
|
| 1014 |
+
"loss": 2.8511,
|
| 1015 |
+
"theoretical_loss": 3.9578312168801597,
|
| 1016 |
+
"tokens_seen": 452198400
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.16,
|
| 1020 |
+
"learning_rate": 0.0005920772649895971,
|
| 1021 |
+
"loss": 2.851,
|
| 1022 |
+
"theoretical_loss": 3.954702950008308,
|
| 1023 |
+
"tokens_seen": 455475200
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.16,
|
| 1027 |
+
"learning_rate": 0.0005912497635710232,
|
| 1028 |
+
"loss": 2.835,
|
| 1029 |
+
"theoretical_loss": 3.9516033583714734,
|
| 1030 |
+
"tokens_seen": 458752000
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.17,
|
| 1034 |
+
"learning_rate": 0.0005904222621524494,
|
| 1035 |
+
"loss": 2.8607,
|
| 1036 |
+
"theoretical_loss": 3.9485319770934355,
|
| 1037 |
+
"tokens_seen": 462028800
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.17,
|
| 1041 |
+
"learning_rate": 0.0005895947607338756,
|
| 1042 |
+
"loss": 2.8461,
|
| 1043 |
+
"theoretical_loss": 3.945488352063391,
|
| 1044 |
+
"tokens_seen": 465305600
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.17,
|
| 1048 |
+
"learning_rate": 0.0005887672593153017,
|
| 1049 |
+
"loss": 2.8307,
|
| 1050 |
+
"theoretical_loss": 3.942472039612926,
|
| 1051 |
+
"tokens_seen": 468582400
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.17,
|
| 1055 |
+
"learning_rate": 0.0005879397578967278,
|
| 1056 |
+
"loss": 2.8269,
|
| 1057 |
+
"theoretical_loss": 3.939482606204863,
|
| 1058 |
+
"tokens_seen": 471859200
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.17,
|
| 1062 |
+
"learning_rate": 0.0005871122564781539,
|
| 1063 |
+
"loss": 2.8407,
|
| 1064 |
+
"theoretical_loss": 3.936519628133466,
|
| 1065 |
+
"tokens_seen": 475136000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.17,
|
| 1069 |
+
"learning_rate": 0.00058628475505958,
|
| 1070 |
+
"loss": 2.8754,
|
| 1071 |
+
"theoretical_loss": 3.9335826912355114,
|
| 1072 |
+
"tokens_seen": 478412800
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.17,
|
| 1076 |
+
"learning_rate": 0.0005854572536410061,
|
| 1077 |
+
"loss": 2.9168,
|
| 1078 |
+
"theoretical_loss": 3.93067139061177,
|
| 1079 |
+
"tokens_seen": 481689600
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.17,
|
| 1083 |
+
"learning_rate": 0.0005846297522224324,
|
| 1084 |
+
"loss": 2.8452,
|
| 1085 |
+
"theoretical_loss": 3.927785330358441,
|
| 1086 |
+
"tokens_seen": 484966400
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.17,
|
| 1090 |
+
"learning_rate": 0.0005838022508038585,
|
| 1091 |
+
"loss": 2.8359,
|
| 1092 |
+
"theoretical_loss": 3.9249241233081333,
|
| 1093 |
+
"tokens_seen": 488243200
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.18,
|
| 1097 |
+
"objective/train/avg_token_score": 0.004644907079637051,
|
| 1098 |
+
"objective/train/avg_weight": 0.995009183883667,
|
| 1099 |
+
"objective/train/docs_used": 287192,
|
| 1100 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 1101 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 1102 |
+
"objective/train/original_loss": 3.014026165008545,
|
| 1103 |
+
"objective/train/std_weight": 0.02483248896896839,
|
| 1104 |
+
"objective/train/theoretical_loss": 3.92208739077998,
|
| 1105 |
+
"objective/train/tokens_used": 511980000,
|
| 1106 |
+
"theoretical_loss": 3.92208739077998,
|
| 1107 |
+
"tokens_seen": 491520000
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.18,
|
| 1111 |
+
"learning_rate": 0.0005829747493852846,
|
| 1112 |
+
"loss": 2.8155,
|
| 1113 |
+
"theoretical_loss": 3.92208739077998,
|
| 1114 |
+
"tokens_seen": 491520000
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.18,
|
| 1118 |
+
"learning_rate": 0.0005821472479667107,
|
| 1119 |
+
"loss": 2.8248,
|
| 1120 |
+
"theoretical_loss": 3.919274762338519,
|
| 1121 |
+
"tokens_seen": 494796800
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.18,
|
| 1125 |
+
"learning_rate": 0.000581319746548137,
|
| 1126 |
+
"loss": 2.8481,
|
| 1127 |
+
"theoretical_loss": 3.9164858755609613,
|
| 1128 |
+
"tokens_seen": 498073600
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.18,
|
| 1132 |
+
"learning_rate": 0.0005804922451295631,
|
| 1133 |
+
"loss": 2.8296,
|
| 1134 |
+
"theoretical_loss": 3.9137203758125176,
|
| 1135 |
+
"tokens_seen": 501350400
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.18,
|
| 1139 |
+
"learning_rate": 0.0005796647437109892,
|
| 1140 |
+
"loss": 2.7929,
|
| 1141 |
+
"theoretical_loss": 3.910977916029439,
|
| 1142 |
+
"tokens_seen": 504627200
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.18,
|
| 1146 |
+
"learning_rate": 0.0005788372422924154,
|
| 1147 |
+
"loss": 2.8029,
|
| 1148 |
+
"theoretical_loss": 3.908258156509472,
|
| 1149 |
+
"tokens_seen": 507904000
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.18,
|
| 1153 |
+
"learning_rate": 0.0005780097408738415,
|
| 1154 |
+
"loss": 2.7949,
|
| 1155 |
+
"theoretical_loss": 3.905560764709417,
|
| 1156 |
+
"tokens_seen": 511180800
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.18,
|
| 1160 |
+
"learning_rate": 0.0005771822394552677,
|
| 1161 |
+
"loss": 2.7728,
|
| 1162 |
+
"theoretical_loss": 3.9028854150495143,
|
| 1163 |
+
"tokens_seen": 514457600
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 0.18,
|
| 1167 |
+
"learning_rate": 0.0005763547380366938,
|
| 1168 |
+
"loss": 2.7991,
|
| 1169 |
+
"theoretical_loss": 3.9002317887243834,
|
| 1170 |
+
"tokens_seen": 517734400
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.19,
|
| 1174 |
+
"learning_rate": 0.0005755272366181199,
|
| 1175 |
+
"loss": 2.7956,
|
| 1176 |
+
"theoretical_loss": 3.897599573520247,
|
| 1177 |
+
"tokens_seen": 521011200
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.19,
|
| 1181 |
+
"learning_rate": 0.0005746997351995461,
|
| 1182 |
+
"loss": 2.8261,
|
| 1183 |
+
"theoretical_loss": 3.8949884636382106,
|
| 1184 |
+
"tokens_seen": 524288000
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.19,
|
| 1188 |
+
"learning_rate": 0.0005738722337809722,
|
| 1189 |
+
"loss": 2.8118,
|
| 1190 |
+
"theoretical_loss": 3.892398159523345,
|
| 1191 |
+
"tokens_seen": 527564800
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.19,
|
| 1195 |
+
"learning_rate": 0.0005730447323623984,
|
| 1196 |
+
"loss": 2.8643,
|
| 1197 |
+
"theoretical_loss": 3.889828367699349,
|
| 1198 |
+
"tokens_seen": 530841600
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.19,
|
| 1202 |
+
"learning_rate": 0.0005722172309438245,
|
| 1203 |
+
"loss": 2.8021,
|
| 1204 |
+
"theoretical_loss": 3.8872788006085894,
|
| 1205 |
+
"tokens_seen": 534118400
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.19,
|
| 1209 |
+
"learning_rate": 0.0005713897295252506,
|
| 1210 |
+
"loss": 2.8285,
|
| 1211 |
+
"theoretical_loss": 3.8847491764572926,
|
| 1212 |
+
"tokens_seen": 537395200
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 0.19,
|
| 1216 |
+
"learning_rate": 0.0005705622281066767,
|
| 1217 |
+
"loss": 2.8119,
|
| 1218 |
+
"theoretical_loss": 3.882239219065708,
|
| 1219 |
+
"tokens_seen": 540672000
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.19,
|
| 1223 |
+
"learning_rate": 0.0005697347266881029,
|
| 1224 |
+
"loss": 2.8372,
|
| 1225 |
+
"theoretical_loss": 3.879748657723039,
|
| 1226 |
+
"tokens_seen": 543948800
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.2,
|
| 1230 |
+
"learning_rate": 0.0005689072252695291,
|
| 1231 |
+
"loss": 2.8645,
|
| 1232 |
+
"theoretical_loss": 3.8772772270469824,
|
| 1233 |
+
"tokens_seen": 547225600
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.2,
|
| 1237 |
+
"learning_rate": 0.0005680797238509552,
|
| 1238 |
+
"loss": 2.8276,
|
| 1239 |
+
"theoretical_loss": 3.8748246668476827,
|
| 1240 |
+
"tokens_seen": 550502400
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.2,
|
| 1244 |
+
"learning_rate": 0.0005672522224323813,
|
| 1245 |
+
"loss": 2.8475,
|
| 1246 |
+
"theoretical_loss": 3.8723907219959486,
|
| 1247 |
+
"tokens_seen": 553779200
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 0.2,
|
| 1251 |
+
"learning_rate": 0.0005664247210138074,
|
| 1252 |
+
"loss": 2.8644,
|
| 1253 |
+
"theoretical_loss": 3.869975142295573,
|
| 1254 |
+
"tokens_seen": 557056000
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.2,
|
| 1258 |
+
"learning_rate": 0.0005655972195952335,
|
| 1259 |
+
"loss": 2.9145,
|
| 1260 |
+
"theoretical_loss": 3.8675776823595998,
|
| 1261 |
+
"tokens_seen": 560332800
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.2,
|
| 1265 |
+
"learning_rate": 0.0005647697181766596,
|
| 1266 |
+
"loss": 2.8816,
|
| 1267 |
+
"theoretical_loss": 3.8651981014904027,
|
| 1268 |
+
"tokens_seen": 563609600
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.2,
|
| 1272 |
+
"learning_rate": 0.0005639422167580859,
|
| 1273 |
+
"loss": 2.8532,
|
| 1274 |
+
"theoretical_loss": 3.8628361635634265,
|
| 1275 |
+
"tokens_seen": 566886400
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 0.2,
|
| 1279 |
+
"learning_rate": 0.0005631312653678835,
|
| 1280 |
+
"loss": 2.8788,
|
| 1281 |
+
"theoretical_loss": 3.8604916369144666,
|
| 1282 |
+
"tokens_seen": 570163200
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.2,
|
| 1286 |
+
"learning_rate": 0.0005623037639493096,
|
| 1287 |
+
"loss": 2.8551,
|
| 1288 |
+
"theoretical_loss": 3.858164294230354,
|
| 1289 |
+
"tokens_seen": 573440000
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.21,
|
| 1293 |
+
"learning_rate": 0.0005614762625307357,
|
| 1294 |
+
"loss": 2.8309,
|
| 1295 |
+
"theoretical_loss": 3.85585391244293,
|
| 1296 |
+
"tokens_seen": 576716800
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.21,
|
| 1300 |
+
"learning_rate": 0.0005606487611121619,
|
| 1301 |
+
"loss": 2.8414,
|
| 1302 |
+
"theoretical_loss": 3.8535602726261864,
|
| 1303 |
+
"tokens_seen": 579993600
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 0.21,
|
| 1307 |
+
"learning_rate": 0.000559821259693588,
|
| 1308 |
+
"loss": 2.8235,
|
| 1309 |
+
"theoretical_loss": 3.851283159896468,
|
| 1310 |
+
"tokens_seen": 583270400
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.21,
|
| 1314 |
+
"learning_rate": 0.0005589937582750142,
|
| 1315 |
+
"loss": 2.8092,
|
| 1316 |
+
"theoretical_loss": 3.8490223633156173,
|
| 1317 |
+
"tokens_seen": 586547200
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 0.21,
|
| 1321 |
+
"learning_rate": 0.0005581662568564403,
|
| 1322 |
+
"loss": 2.7915,
|
| 1323 |
+
"theoretical_loss": 3.846777675796974,
|
| 1324 |
+
"tokens_seen": 589824000
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.21,
|
| 1328 |
+
"learning_rate": 0.0005573387554378664,
|
| 1329 |
+
"loss": 2.7981,
|
| 1330 |
+
"theoretical_loss": 3.844548894014116,
|
| 1331 |
+
"tokens_seen": 593100800
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 0.21,
|
| 1335 |
+
"learning_rate": 0.0005565112540192926,
|
| 1336 |
+
"loss": 2.8022,
|
| 1337 |
+
"theoretical_loss": 3.8423358183122582,
|
| 1338 |
+
"tokens_seen": 596377600
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.21,
|
| 1342 |
+
"learning_rate": 0.0005556837526007187,
|
| 1343 |
+
"loss": 2.7754,
|
| 1344 |
+
"theoretical_loss": 3.840138252622208,
|
| 1345 |
+
"tokens_seen": 599654400
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.22,
|
| 1349 |
+
"learning_rate": 0.0005548562511821448,
|
| 1350 |
+
"loss": 2.8037,
|
| 1351 |
+
"theoretical_loss": 3.837956004376799,
|
| 1352 |
+
"tokens_seen": 602931200
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.22,
|
| 1356 |
+
"learning_rate": 0.000554028749763571,
|
| 1357 |
+
"loss": 2.7709,
|
| 1358 |
+
"theoretical_loss": 3.8357888844297094,
|
| 1359 |
+
"tokens_seen": 606208000
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.22,
|
| 1363 |
+
"learning_rate": 0.0005532012483449971,
|
| 1364 |
+
"loss": 2.7927,
|
| 1365 |
+
"theoretical_loss": 3.8336367069765958,
|
| 1366 |
+
"tokens_seen": 609484800
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.22,
|
| 1370 |
+
"learning_rate": 0.0005523737469264233,
|
| 1371 |
+
"loss": 2.7451,
|
| 1372 |
+
"theoretical_loss": 3.8314992894784536,
|
| 1373 |
+
"tokens_seen": 612761600
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 0.22,
|
| 1377 |
+
"learning_rate": 0.0005515462455078494,
|
| 1378 |
+
"loss": 2.7896,
|
| 1379 |
+
"theoretical_loss": 3.829376452587134,
|
| 1380 |
+
"tokens_seen": 616038400
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.22,
|
| 1384 |
+
"learning_rate": 0.0005507187440892756,
|
| 1385 |
+
"loss": 2.8047,
|
| 1386 |
+
"theoretical_loss": 3.827268020072948,
|
| 1387 |
+
"tokens_seen": 619315200
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 0.22,
|
| 1391 |
+
"learning_rate": 0.0005498912426707017,
|
| 1392 |
+
"loss": 2.8391,
|
| 1393 |
+
"theoretical_loss": 3.8251738187542843,
|
| 1394 |
+
"tokens_seen": 622592000
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.22,
|
| 1398 |
+
"learning_rate": 0.0005490637412521278,
|
| 1399 |
+
"loss": 2.8195,
|
| 1400 |
+
"theoretical_loss": 3.8230936784291787,
|
| 1401 |
+
"tokens_seen": 625868800
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.22,
|
| 1405 |
+
"learning_rate": 0.0005482362398335541,
|
| 1406 |
+
"loss": 2.8531,
|
| 1407 |
+
"theoretical_loss": 3.8210274318087656,
|
| 1408 |
+
"tokens_seen": 629145600
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.23,
|
| 1412 |
+
"learning_rate": 0.0005474087384149802,
|
| 1413 |
+
"loss": 2.8001,
|
| 1414 |
+
"theoretical_loss": 3.818974914452557,
|
| 1415 |
+
"tokens_seen": 632422400
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 0.23,
|
| 1419 |
+
"learning_rate": 0.0005465812369964063,
|
| 1420 |
+
"loss": 2.7996,
|
| 1421 |
+
"theoretical_loss": 3.8169359647054835,
|
| 1422 |
+
"tokens_seen": 635699200
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 0.23,
|
| 1426 |
+
"learning_rate": 0.0005457537355778324,
|
| 1427 |
+
"loss": 2.7817,
|
| 1428 |
+
"theoretical_loss": 3.8149104236366433,
|
| 1429 |
+
"tokens_seen": 638976000
|
| 1430 |
+
},
|
| 1431 |
+
{
|
| 1432 |
+
"epoch": 0.23,
|
| 1433 |
+
"learning_rate": 0.0005449262341592585,
|
| 1434 |
+
"loss": 2.7613,
|
| 1435 |
+
"theoretical_loss": 3.8128981349797098,
|
| 1436 |
+
"tokens_seen": 642252800
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.23,
|
| 1440 |
+
"learning_rate": 0.0005440987327406847,
|
| 1441 |
+
"loss": 2.7851,
|
| 1442 |
+
"theoretical_loss": 3.8108989450749293,
|
| 1443 |
+
"tokens_seen": 645529600
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 0.23,
|
| 1447 |
+
"learning_rate": 0.0005432712313221109,
|
| 1448 |
+
"loss": 2.7738,
|
| 1449 |
+
"theoretical_loss": 3.8089127028126764,
|
| 1450 |
+
"tokens_seen": 648806400
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.23,
|
| 1454 |
+
"learning_rate": 0.000542443729903537,
|
| 1455 |
+
"loss": 2.8114,
|
| 1456 |
+
"theoretical_loss": 3.8069392595785083,
|
| 1457 |
+
"tokens_seen": 652083200
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"debugging/Self-BLEU-5": 0.5265375629586004,
|
| 1461 |
+
"debugging/distinct-1-grams": 0.7435820408094715,
|
| 1462 |
+
"debugging/distinct-2-grams": 0.9558103821233092,
|
| 1463 |
+
"debugging/entropy-1-grams": 5.931434510687563,
|
| 1464 |
+
"debugging/entropy-2-grams": 6.886416755326388,
|
| 1465 |
+
"debugging/length": 521.9230769230769,
|
| 1466 |
+
"debugging/num_segments": 13,
|
| 1467 |
+
"epoch": 0.23,
|
| 1468 |
+
"objective/train/avg_token_score": 0.022742915898561478,
|
| 1469 |
+
"objective/train/avg_weight": 0.9755392670631409,
|
| 1470 |
+
"objective/train/docs_used": 379091,
|
| 1471 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 1472 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 1473 |
+
"objective/train/original_loss": 3.0632386207580566,
|
| 1474 |
+
"objective/train/std_weight": 0.08433963358402252,
|
| 1475 |
+
"objective/train/theoretical_loss": 3.804978469199669,
|
| 1476 |
+
"objective/train/tokens_used": 675820000,
|
| 1477 |
+
"theoretical_loss": 3.804978469199669,
|
| 1478 |
+
"tokens_seen": 655360000
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.23,
|
| 1482 |
+
"learning_rate": 0.0005416162284849631,
|
| 1483 |
+
"loss": 2.805,
|
| 1484 |
+
"theoretical_loss": 3.804978469199669,
|
| 1485 |
+
"tokens_seen": 655360000
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 0.24,
|
| 1489 |
+
"learning_rate": 0.0005407887270663892,
|
| 1490 |
+
"loss": 2.8212,
|
| 1491 |
+
"theoretical_loss": 3.803030187893005,
|
| 1492 |
+
"tokens_seen": 658636800
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 0.24,
|
| 1496 |
+
"learning_rate": 0.0005399612256478154,
|
| 1497 |
+
"loss": 2.82,
|
| 1498 |
+
"theoretical_loss": 3.8010942742142415,
|
| 1499 |
+
"tokens_seen": 661913600
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 0.24,
|
| 1503 |
+
"learning_rate": 0.0005391337242292415,
|
| 1504 |
+
"loss": 2.8474,
|
| 1505 |
+
"theoretical_loss": 3.799170589008585,
|
| 1506 |
+
"tokens_seen": 665190400
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 0.24,
|
| 1510 |
+
"learning_rate": 0.0005383062228106677,
|
| 1511 |
+
"loss": 2.8322,
|
| 1512 |
+
"theoretical_loss": 3.7972589953626006,
|
| 1513 |
+
"tokens_seen": 668467200
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.24,
|
| 1517 |
+
"learning_rate": 0.0005374787213920938,
|
| 1518 |
+
"loss": 2.8432,
|
| 1519 |
+
"theoretical_loss": 3.795359358557337,
|
| 1520 |
+
"tokens_seen": 671744000
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.24,
|
| 1524 |
+
"learning_rate": 0.0005366512199735199,
|
| 1525 |
+
"loss": 2.7784,
|
| 1526 |
+
"theoretical_loss": 3.79347154602265,
|
| 1527 |
+
"tokens_seen": 675020800
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 0.24,
|
| 1531 |
+
"learning_rate": 0.0005358237185549461,
|
| 1532 |
+
"loss": 2.824,
|
| 1533 |
+
"theoretical_loss": 3.7915954272926955,
|
| 1534 |
+
"tokens_seen": 678297600
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 0.24,
|
| 1538 |
+
"learning_rate": 0.0005349962171363722,
|
| 1539 |
+
"loss": 2.7824,
|
| 1540 |
+
"theoretical_loss": 3.789730873962557,
|
| 1541 |
+
"tokens_seen": 681574400
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 0.24,
|
| 1545 |
+
"learning_rate": 0.0005341687157177983,
|
| 1546 |
+
"loss": 2.7185,
|
| 1547 |
+
"theoretical_loss": 3.787877759645963,
|
| 1548 |
+
"tokens_seen": 684851200
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 0.25,
|
| 1552 |
+
"learning_rate": 0.0005333412142992244,
|
| 1553 |
+
"loss": 2.7511,
|
| 1554 |
+
"theoretical_loss": 3.7860359599340776,
|
| 1555 |
+
"tokens_seen": 688128000
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 0.25,
|
| 1559 |
+
"learning_rate": 0.0005325137128806506,
|
| 1560 |
+
"loss": 2.791,
|
| 1561 |
+
"theoretical_loss": 3.784205352355321,
|
| 1562 |
+
"tokens_seen": 691404800
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.25,
|
| 1566 |
+
"learning_rate": 0.0005316862114620768,
|
| 1567 |
+
"loss": 2.8214,
|
| 1568 |
+
"theoretical_loss": 3.782385816336189,
|
| 1569 |
+
"tokens_seen": 694681600
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 0.25,
|
| 1573 |
+
"learning_rate": 0.0005308587100435029,
|
| 1574 |
+
"loss": 2.7788,
|
| 1575 |
+
"theoretical_loss": 3.7805772331630516,
|
| 1576 |
+
"tokens_seen": 697958400
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 0.25,
|
| 1580 |
+
"learning_rate": 0.0005300312086249291,
|
| 1581 |
+
"loss": 2.7553,
|
| 1582 |
+
"theoretical_loss": 3.7787794859448898,
|
| 1583 |
+
"tokens_seen": 701235200
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 0.25,
|
| 1587 |
+
"learning_rate": 0.0005292037072063552,
|
| 1588 |
+
"loss": 2.7382,
|
| 1589 |
+
"theoretical_loss": 3.7769924595769546,
|
| 1590 |
+
"tokens_seen": 704512000
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 0.25,
|
| 1594 |
+
"learning_rate": 0.0005283762057877813,
|
| 1595 |
+
"loss": 2.7255,
|
| 1596 |
+
"theoretical_loss": 3.7752160407053115,
|
| 1597 |
+
"tokens_seen": 707788800
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 0.25,
|
| 1601 |
+
"learning_rate": 0.0005275487043692075,
|
| 1602 |
+
"loss": 2.7101,
|
| 1603 |
+
"theoretical_loss": 3.7734501176922493,
|
| 1604 |
+
"tokens_seen": 711065600
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 0.26,
|
| 1608 |
+
"learning_rate": 0.0005267212029506337,
|
| 1609 |
+
"loss": 2.7514,
|
| 1610 |
+
"theoretical_loss": 3.7716945805825337,
|
| 1611 |
+
"tokens_seen": 714342400
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 0.26,
|
| 1615 |
+
"learning_rate": 0.0005258937015320598,
|
| 1616 |
+
"loss": 2.7886,
|
| 1617 |
+
"theoretical_loss": 3.7699493210704667,
|
| 1618 |
+
"tokens_seen": 717619200
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 0.26,
|
| 1622 |
+
"learning_rate": 0.0005250662001134859,
|
| 1623 |
+
"loss": 2.7644,
|
| 1624 |
+
"theoretical_loss": 3.7682142324677455,
|
| 1625 |
+
"tokens_seen": 720896000
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 0.26,
|
| 1629 |
+
"learning_rate": 0.000524238698694912,
|
| 1630 |
+
"loss": 2.8249,
|
| 1631 |
+
"theoretical_loss": 3.7664892096720886,
|
| 1632 |
+
"tokens_seen": 724172800
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 0.26,
|
| 1636 |
+
"learning_rate": 0.0005234111972763382,
|
| 1637 |
+
"loss": 2.7615,
|
| 1638 |
+
"theoretical_loss": 3.7647741491366067,
|
| 1639 |
+
"tokens_seen": 727449600
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 0.26,
|
| 1643 |
+
"learning_rate": 0.0005225836958577644,
|
| 1644 |
+
"loss": 2.7764,
|
| 1645 |
+
"theoretical_loss": 3.7630689488399027,
|
| 1646 |
+
"tokens_seen": 730726400
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 0.26,
|
| 1650 |
+
"learning_rate": 0.0005217561944391905,
|
| 1651 |
+
"loss": 2.7431,
|
| 1652 |
+
"theoretical_loss": 3.7613735082568764,
|
| 1653 |
+
"tokens_seen": 734003200
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 0.26,
|
| 1657 |
+
"learning_rate": 0.0005209286930206166,
|
| 1658 |
+
"loss": 2.7793,
|
| 1659 |
+
"theoretical_loss": 3.759687728330217,
|
| 1660 |
+
"tokens_seen": 737280000
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 0.26,
|
| 1664 |
+
"learning_rate": 0.0005201011916020427,
|
| 1665 |
+
"loss": 2.7728,
|
| 1666 |
+
"theoretical_loss": 3.75801151144256,
|
| 1667 |
+
"tokens_seen": 740556800
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 0.27,
|
| 1671 |
+
"learning_rate": 0.0005192736901834689,
|
| 1672 |
+
"loss": 2.7565,
|
| 1673 |
+
"theoretical_loss": 3.756344761389295,
|
| 1674 |
+
"tokens_seen": 743833600
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 0.27,
|
| 1678 |
+
"learning_rate": 0.000518446188764895,
|
| 1679 |
+
"loss": 2.7275,
|
| 1680 |
+
"theoretical_loss": 3.754687383352003,
|
| 1681 |
+
"tokens_seen": 747110400
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 0.27,
|
| 1685 |
+
"learning_rate": 0.0005176186873463211,
|
| 1686 |
+
"loss": 2.7306,
|
| 1687 |
+
"theoretical_loss": 3.7530392838725097,
|
| 1688 |
+
"tokens_seen": 750387200
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 0.27,
|
| 1692 |
+
"learning_rate": 0.0005167911859277473,
|
| 1693 |
+
"loss": 2.7465,
|
| 1694 |
+
"theoretical_loss": 3.751400370827529,
|
| 1695 |
+
"tokens_seen": 753664000
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 0.27,
|
| 1699 |
+
"learning_rate": 0.0005159636845091734,
|
| 1700 |
+
"loss": 2.7684,
|
| 1701 |
+
"theoretical_loss": 3.749770553403895,
|
| 1702 |
+
"tokens_seen": 756940800
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 0.27,
|
| 1706 |
+
"learning_rate": 0.0005151361830905996,
|
| 1707 |
+
"loss": 2.7327,
|
| 1708 |
+
"theoretical_loss": 3.748149742074355,
|
| 1709 |
+
"tokens_seen": 760217600
|
| 1710 |
+
},
|
| 1711 |
+
{
|
| 1712 |
+
"epoch": 0.27,
|
| 1713 |
+
"learning_rate": 0.0005143086816720257,
|
| 1714 |
+
"loss": 2.7618,
|
| 1715 |
+
"theoretical_loss": 3.746537848573908,
|
| 1716 |
+
"tokens_seen": 763494400
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 0.27,
|
| 1720 |
+
"learning_rate": 0.0005134811802534518,
|
| 1721 |
+
"loss": 2.7765,
|
| 1722 |
+
"theoretical_loss": 3.744934785876686,
|
| 1723 |
+
"tokens_seen": 766771200
|
| 1724 |
+
},
|
| 1725 |
+
{
|
| 1726 |
+
"epoch": 0.28,
|
| 1727 |
+
"learning_rate": 0.0005126536788348779,
|
| 1728 |
+
"loss": 2.7845,
|
| 1729 |
+
"theoretical_loss": 3.7433404681733475,
|
| 1730 |
+
"tokens_seen": 770048000
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 0.28,
|
| 1734 |
+
"learning_rate": 0.0005118261774163041,
|
| 1735 |
+
"loss": 2.7901,
|
| 1736 |
+
"theoretical_loss": 3.7417548108489846,
|
| 1737 |
+
"tokens_seen": 773324800
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 0.28,
|
| 1741 |
+
"learning_rate": 0.0005109986759977303,
|
| 1742 |
+
"loss": 2.7636,
|
| 1743 |
+
"theoretical_loss": 3.740177730461517,
|
| 1744 |
+
"tokens_seen": 776601600
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 0.28,
|
| 1748 |
+
"learning_rate": 0.0005101711745791564,
|
| 1749 |
+
"loss": 2.8129,
|
| 1750 |
+
"theoretical_loss": 3.73860914472057,
|
| 1751 |
+
"tokens_seen": 779878400
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 0.28,
|
| 1755 |
+
"learning_rate": 0.0005093436731605825,
|
| 1756 |
+
"loss": 2.7786,
|
| 1757 |
+
"theoretical_loss": 3.7370489724668197,
|
| 1758 |
+
"tokens_seen": 783155200
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 0.28,
|
| 1762 |
+
"learning_rate": 0.0005085161717420087,
|
| 1763 |
+
"loss": 2.7891,
|
| 1764 |
+
"theoretical_loss": 3.735497133651788,
|
| 1765 |
+
"tokens_seen": 786432000
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 0.28,
|
| 1769 |
+
"learning_rate": 0.0005076886703234348,
|
| 1770 |
+
"loss": 2.7787,
|
| 1771 |
+
"theoretical_loss": 3.733953549318091,
|
| 1772 |
+
"tokens_seen": 789708800
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 0.28,
|
| 1776 |
+
"learning_rate": 0.000506861168904861,
|
| 1777 |
+
"loss": 2.7923,
|
| 1778 |
+
"theoretical_loss": 3.7324181415801094,
|
| 1779 |
+
"tokens_seen": 792985600
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 0.28,
|
| 1783 |
+
"learning_rate": 0.0005060336674862872,
|
| 1784 |
+
"loss": 2.7873,
|
| 1785 |
+
"theoretical_loss": 3.7308908336050814,
|
| 1786 |
+
"tokens_seen": 796262400
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 0.29,
|
| 1790 |
+
"learning_rate": 0.0005052061660677133,
|
| 1791 |
+
"loss": 2.772,
|
| 1792 |
+
"theoretical_loss": 3.729371549594614,
|
| 1793 |
+
"tokens_seen": 799539200
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 0.29,
|
| 1797 |
+
"learning_rate": 0.0005043786646491394,
|
| 1798 |
+
"loss": 2.7862,
|
| 1799 |
+
"theoretical_loss": 3.7278602147665776,
|
| 1800 |
+
"tokens_seen": 802816000
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 0.29,
|
| 1804 |
+
"learning_rate": 0.0005035511632305655,
|
| 1805 |
+
"loss": 2.7841,
|
| 1806 |
+
"theoretical_loss": 3.726356755337407,
|
| 1807 |
+
"tokens_seen": 806092800
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 0.29,
|
| 1811 |
+
"learning_rate": 0.0005027236618119917,
|
| 1812 |
+
"loss": 2.7735,
|
| 1813 |
+
"theoretical_loss": 3.724861098504767,
|
| 1814 |
+
"tokens_seen": 809369600
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.29,
|
| 1818 |
+
"learning_rate": 0.0005018961603934179,
|
| 1819 |
+
"loss": 2.7389,
|
| 1820 |
+
"theoretical_loss": 3.7233731724305974,
|
| 1821 |
+
"tokens_seen": 812646400
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 0.29,
|
| 1825 |
+
"learning_rate": 0.000501068658974844,
|
| 1826 |
+
"loss": 2.7629,
|
| 1827 |
+
"theoretical_loss": 3.7218929062245105,
|
| 1828 |
+
"tokens_seen": 815923200
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 0.29,
|
| 1832 |
+
"objective/train/avg_token_score": 0.009068925864994526,
|
| 1833 |
+
"objective/train/avg_weight": 0.9902427196502686,
|
| 1834 |
+
"objective/train/docs_used": 471128,
|
| 1835 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 1836 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 1837 |
+
"objective/train/original_loss": 2.6760623455047607,
|
| 1838 |
+
"objective/train/std_weight": 0.06257461756467819,
|
| 1839 |
+
"objective/train/theoretical_loss": 3.7204202299275475,
|
| 1840 |
+
"objective/train/tokens_used": 839660000,
|
| 1841 |
+
"theoretical_loss": 3.7204202299275475,
|
| 1842 |
+
"tokens_seen": 819200000
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 0.29,
|
| 1846 |
+
"learning_rate": 0.0005002411575562701,
|
| 1847 |
+
"loss": 2.7057,
|
| 1848 |
+
"theoretical_loss": 3.7204202299275475,
|
| 1849 |
+
"tokens_seen": 819200000
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.29,
|
| 1853 |
+
"learning_rate": 0.0004994136561376962,
|
| 1854 |
+
"loss": 2.7306,
|
| 1855 |
+
"theoretical_loss": 3.7189550744962707,
|
| 1856 |
+
"tokens_seen": 822476800
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.29,
|
| 1860 |
+
"learning_rate": 0.0004985861547191224,
|
| 1861 |
+
"loss": 2.7091,
|
| 1862 |
+
"theoretical_loss": 3.717497371787192,
|
| 1863 |
+
"tokens_seen": 825753600
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 0.3,
|
| 1867 |
+
"learning_rate": 0.0004977586533005485,
|
| 1868 |
+
"loss": 2.7075,
|
| 1869 |
+
"theoretical_loss": 3.7160470545415274,
|
| 1870 |
+
"tokens_seen": 829030400
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 0.3,
|
| 1874 |
+
"learning_rate": 0.0004969311518819746,
|
| 1875 |
+
"loss": 2.7036,
|
| 1876 |
+
"theoretical_loss": 3.714604056370267,
|
| 1877 |
+
"tokens_seen": 832307200
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 0.3,
|
| 1881 |
+
"learning_rate": 0.0004961036504634008,
|
| 1882 |
+
"loss": 2.716,
|
| 1883 |
+
"theoretical_loss": 3.713168311739558,
|
| 1884 |
+
"tokens_seen": 835584000
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 0.3,
|
| 1888 |
+
"learning_rate": 0.0004952761490448269,
|
| 1889 |
+
"loss": 2.7011,
|
| 1890 |
+
"theoretical_loss": 3.7117397559563843,
|
| 1891 |
+
"tokens_seen": 838860800
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 0.3,
|
| 1895 |
+
"learning_rate": 0.0004944486476262531,
|
| 1896 |
+
"loss": 2.6838,
|
| 1897 |
+
"theoretical_loss": 3.710318325154545,
|
| 1898 |
+
"tokens_seen": 842137600
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 0.3,
|
| 1902 |
+
"learning_rate": 0.0004936211462076792,
|
| 1903 |
+
"loss": 2.7304,
|
| 1904 |
+
"theoretical_loss": 3.7089039562809223,
|
| 1905 |
+
"tokens_seen": 845414400
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 0.3,
|
| 1909 |
+
"learning_rate": 0.0004927936447891053,
|
| 1910 |
+
"loss": 2.7027,
|
| 1911 |
+
"theoretical_loss": 3.7074965870820193,
|
| 1912 |
+
"tokens_seen": 848691200
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 0.3,
|
| 1916 |
+
"learning_rate": 0.0004919661433705314,
|
| 1917 |
+
"loss": 2.7073,
|
| 1918 |
+
"theoretical_loss": 3.7060961560907857,
|
| 1919 |
+
"tokens_seen": 851968000
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 0.31,
|
| 1923 |
+
"learning_rate": 0.0004911386419519576,
|
| 1924 |
+
"loss": 2.7664,
|
| 1925 |
+
"theoretical_loss": 3.7047026026137,
|
| 1926 |
+
"tokens_seen": 855244800
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 0.31,
|
| 1930 |
+
"learning_rate": 0.0004903111405333838,
|
| 1931 |
+
"loss": 2.7282,
|
| 1932 |
+
"theoretical_loss": 3.7033158667181154,
|
| 1933 |
+
"tokens_seen": 858521600
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 0.31,
|
| 1937 |
+
"learning_rate": 0.0004894836391148099,
|
| 1938 |
+
"loss": 2.7546,
|
| 1939 |
+
"theoretical_loss": 3.701935889219863,
|
| 1940 |
+
"tokens_seen": 861798400
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.31,
|
| 1944 |
+
"learning_rate": 0.000488656137696236,
|
| 1945 |
+
"loss": 2.7233,
|
| 1946 |
+
"theoretical_loss": 3.7005626116710966,
|
| 1947 |
+
"tokens_seen": 865075200
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 0.31,
|
| 1951 |
+
"learning_rate": 0.00048782863627766217,
|
| 1952 |
+
"loss": 2.7292,
|
| 1953 |
+
"theoretical_loss": 3.69919597634839,
|
| 1954 |
+
"tokens_seen": 868352000
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 0.31,
|
| 1958 |
+
"learning_rate": 0.0004870011348590883,
|
| 1959 |
+
"loss": 2.7075,
|
| 1960 |
+
"theoretical_loss": 3.6978359262410603,
|
| 1961 |
+
"tokens_seen": 871628800
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 0.31,
|
| 1965 |
+
"learning_rate": 0.0004861736334405145,
|
| 1966 |
+
"loss": 2.7358,
|
| 1967 |
+
"theoretical_loss": 3.6964824050397276,
|
| 1968 |
+
"tokens_seen": 874905600
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 0.31,
|
| 1972 |
+
"learning_rate": 0.00048534613202194063,
|
| 1973 |
+
"loss": 2.676,
|
| 1974 |
+
"theoretical_loss": 3.6951353571251015,
|
| 1975 |
+
"tokens_seen": 878182400
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 0.31,
|
| 1979 |
+
"learning_rate": 0.00048451863060336675,
|
| 1980 |
+
"loss": 2.6779,
|
| 1981 |
+
"theoretical_loss": 3.693794727556988,
|
| 1982 |
+
"tokens_seen": 881459200
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.32,
|
| 1986 |
+
"learning_rate": 0.00048369112918479287,
|
| 1987 |
+
"loss": 2.6825,
|
| 1988 |
+
"theoretical_loss": 3.692460462063506,
|
| 1989 |
+
"tokens_seen": 884736000
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 0.32,
|
| 1993 |
+
"learning_rate": 0.000482863627766219,
|
| 1994 |
+
"loss": 2.7369,
|
| 1995 |
+
"theoretical_loss": 3.691132507030521,
|
| 1996 |
+
"tokens_seen": 888012800
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 0.32,
|
| 2000 |
+
"learning_rate": 0.00048203612634764516,
|
| 2001 |
+
"loss": 2.7514,
|
| 2002 |
+
"theoretical_loss": 3.6898108094912816,
|
| 2003 |
+
"tokens_seen": 891289600
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 0.32,
|
| 2007 |
+
"learning_rate": 0.0004812086249290713,
|
| 2008 |
+
"loss": 2.7082,
|
| 2009 |
+
"theoretical_loss": 3.6884953171162556,
|
| 2010 |
+
"tokens_seen": 894566400
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 0.32,
|
| 2014 |
+
"learning_rate": 0.00048038112351049745,
|
| 2015 |
+
"loss": 2.7144,
|
| 2016 |
+
"theoretical_loss": 3.6871859782031624,
|
| 2017 |
+
"tokens_seen": 897843200
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 0.32,
|
| 2021 |
+
"learning_rate": 0.0004795536220919236,
|
| 2022 |
+
"loss": 2.7252,
|
| 2023 |
+
"theoretical_loss": 3.685882741667202,
|
| 2024 |
+
"tokens_seen": 901120000
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 0.32,
|
| 2028 |
+
"learning_rate": 0.00047872612067334974,
|
| 2029 |
+
"loss": 2.7262,
|
| 2030 |
+
"theoretical_loss": 3.684585557031461,
|
| 2031 |
+
"tokens_seen": 904396800
|
| 2032 |
+
},
|
| 2033 |
+
{
|
| 2034 |
+
"epoch": 0.32,
|
| 2035 |
+
"learning_rate": 0.00047789861925477586,
|
| 2036 |
+
"loss": 2.7381,
|
| 2037 |
+
"theoretical_loss": 3.6832943744175126,
|
| 2038 |
+
"tokens_seen": 907673600
|
| 2039 |
+
},
|
| 2040 |
+
{
|
| 2041 |
+
"epoch": 0.33,
|
| 2042 |
+
"learning_rate": 0.000477071117836202,
|
| 2043 |
+
"loss": 2.7141,
|
| 2044 |
+
"theoretical_loss": 3.682009144536188,
|
| 2045 |
+
"tokens_seen": 910950400
|
| 2046 |
+
},
|
| 2047 |
+
{
|
| 2048 |
+
"epoch": 0.33,
|
| 2049 |
+
"learning_rate": 0.0004762436164176281,
|
| 2050 |
+
"loss": 2.7418,
|
| 2051 |
+
"theoretical_loss": 3.680729818678526,
|
| 2052 |
+
"tokens_seen": 914227200
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"epoch": 0.33,
|
| 2056 |
+
"learning_rate": 0.0004754161149990543,
|
| 2057 |
+
"loss": 2.7204,
|
| 2058 |
+
"theoretical_loss": 3.6794563487068936,
|
| 2059 |
+
"tokens_seen": 917504000
|
| 2060 |
+
},
|
| 2061 |
+
{
|
| 2062 |
+
"epoch": 0.33,
|
| 2063 |
+
"learning_rate": 0.00047458861358048044,
|
| 2064 |
+
"loss": 2.7297,
|
| 2065 |
+
"theoretical_loss": 3.6781886870462692,
|
| 2066 |
+
"tokens_seen": 920780800
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 0.33,
|
| 2070 |
+
"learning_rate": 0.00047376111216190656,
|
| 2071 |
+
"loss": 2.7465,
|
| 2072 |
+
"theoretical_loss": 3.676926786675698,
|
| 2073 |
+
"tokens_seen": 924057600
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 0.33,
|
| 2077 |
+
"learning_rate": 0.0004729336107433327,
|
| 2078 |
+
"loss": 2.6863,
|
| 2079 |
+
"theoretical_loss": 3.6756706011198963,
|
| 2080 |
+
"tokens_seen": 927334400
|
| 2081 |
+
},
|
| 2082 |
+
{
|
| 2083 |
+
"epoch": 0.33,
|
| 2084 |
+
"learning_rate": 0.0004721061093247588,
|
| 2085 |
+
"loss": 2.6538,
|
| 2086 |
+
"theoretical_loss": 3.6744200844410217,
|
| 2087 |
+
"tokens_seen": 930611200
|
| 2088 |
+
},
|
| 2089 |
+
{
|
| 2090 |
+
"epoch": 0.33,
|
| 2091 |
+
"learning_rate": 0.00047127860790618497,
|
| 2092 |
+
"loss": 2.6555,
|
| 2093 |
+
"theoretical_loss": 3.6731751912305914,
|
| 2094 |
+
"tokens_seen": 933888000
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"epoch": 0.33,
|
| 2098 |
+
"learning_rate": 0.00047045110648761114,
|
| 2099 |
+
"loss": 2.688,
|
| 2100 |
+
"theoretical_loss": 3.671935876601547,
|
| 2101 |
+
"tokens_seen": 937164800
|
| 2102 |
+
},
|
| 2103 |
+
{
|
| 2104 |
+
"epoch": 0.34,
|
| 2105 |
+
"learning_rate": 0.00046962360506903726,
|
| 2106 |
+
"loss": 2.6286,
|
| 2107 |
+
"theoretical_loss": 3.6707020961804715,
|
| 2108 |
+
"tokens_seen": 940441600
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 0.34,
|
| 2112 |
+
"learning_rate": 0.00046879610365046343,
|
| 2113 |
+
"loss": 2.6828,
|
| 2114 |
+
"theoretical_loss": 3.6694738060999468,
|
| 2115 |
+
"tokens_seen": 943718400
|
| 2116 |
+
},
|
| 2117 |
+
{
|
| 2118 |
+
"epoch": 0.34,
|
| 2119 |
+
"learning_rate": 0.00046796860223188955,
|
| 2120 |
+
"loss": 2.7023,
|
| 2121 |
+
"theoretical_loss": 3.668250962991049,
|
| 2122 |
+
"tokens_seen": 946995200
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 0.34,
|
| 2126 |
+
"learning_rate": 0.00046714110081331567,
|
| 2127 |
+
"loss": 2.7018,
|
| 2128 |
+
"theoretical_loss": 3.667033523975983,
|
| 2129 |
+
"tokens_seen": 950272000
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 0.34,
|
| 2133 |
+
"learning_rate": 0.0004663135993947418,
|
| 2134 |
+
"loss": 2.6876,
|
| 2135 |
+
"theoretical_loss": 3.66582144666085,
|
| 2136 |
+
"tokens_seen": 953548800
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 0.34,
|
| 2140 |
+
"learning_rate": 0.0004654860979761679,
|
| 2141 |
+
"loss": 2.687,
|
| 2142 |
+
"theoretical_loss": 3.664614689128546,
|
| 2143 |
+
"tokens_seen": 956825600
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 0.34,
|
| 2147 |
+
"learning_rate": 0.00046465859655759413,
|
| 2148 |
+
"loss": 2.6865,
|
| 2149 |
+
"theoretical_loss": 3.6634132099317886,
|
| 2150 |
+
"tokens_seen": 960102400
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 0.34,
|
| 2154 |
+
"learning_rate": 0.00046383109513902025,
|
| 2155 |
+
"loss": 2.6943,
|
| 2156 |
+
"theoretical_loss": 3.662216968086267,
|
| 2157 |
+
"tokens_seen": 963379200
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 0.35,
|
| 2161 |
+
"learning_rate": 0.00046300359372044637,
|
| 2162 |
+
"loss": 2.7232,
|
| 2163 |
+
"theoretical_loss": 3.6610259230639217,
|
| 2164 |
+
"tokens_seen": 966656000
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 0.35,
|
| 2168 |
+
"learning_rate": 0.0004621760923018725,
|
| 2169 |
+
"loss": 2.7559,
|
| 2170 |
+
"theoretical_loss": 3.659840034786333,
|
| 2171 |
+
"tokens_seen": 969932800
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 0.35,
|
| 2175 |
+
"learning_rate": 0.00046134859088329866,
|
| 2176 |
+
"loss": 2.8156,
|
| 2177 |
+
"theoretical_loss": 3.6586592636182376,
|
| 2178 |
+
"tokens_seen": 973209600
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 0.35,
|
| 2182 |
+
"learning_rate": 0.0004605210894647248,
|
| 2183 |
+
"loss": 2.8084,
|
| 2184 |
+
"theoretical_loss": 3.6574835703611566,
|
| 2185 |
+
"tokens_seen": 976486400
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 0.35,
|
| 2189 |
+
"learning_rate": 0.00045969358804615095,
|
| 2190 |
+
"loss": 2.793,
|
| 2191 |
+
"theoretical_loss": 3.6563129162471313,
|
| 2192 |
+
"tokens_seen": 979763200
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"debugging/Self-BLEU-5": 0.4286046663919377,
|
| 2196 |
+
"debugging/distinct-1-grams": 0.8147567798871364,
|
| 2197 |
+
"debugging/distinct-2-grams": 0.9823269374342457,
|
| 2198 |
+
"debugging/entropy-1-grams": 6.1671920556004824,
|
| 2199 |
+
"debugging/entropy-2-grams": 6.947028138756313,
|
| 2200 |
+
"debugging/length": 477.53333333333336,
|
| 2201 |
+
"debugging/num_segments": 15,
|
| 2202 |
+
"epoch": 0.35,
|
| 2203 |
+
"objective/train/avg_token_score": 0.020611366257071495,
|
| 2204 |
+
"objective/train/avg_weight": 0.9778223633766174,
|
| 2205 |
+
"objective/train/docs_used": 560408,
|
| 2206 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 2207 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 2208 |
+
"objective/train/original_loss": 2.926975965499878,
|
| 2209 |
+
"objective/train/std_weight": 0.09140268713235855,
|
| 2210 |
+
"objective/train/theoretical_loss": 3.6551472629325787,
|
| 2211 |
+
"objective/train/tokens_used": 1003500000,
|
| 2212 |
+
"theoretical_loss": 3.6551472629325787,
|
| 2213 |
+
"tokens_seen": 983040000
|
| 2214 |
+
},
|
| 2215 |
+
{
|
| 2216 |
+
"epoch": 0.35,
|
| 2217 |
+
"learning_rate": 0.00045886608662757707,
|
| 2218 |
+
"loss": 2.7908,
|
| 2219 |
+
"theoretical_loss": 3.6551472629325787,
|
| 2220 |
+
"tokens_seen": 983040000
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"epoch": 0.35,
|
| 2224 |
+
"learning_rate": 0.00045803858520900324,
|
| 2225 |
+
"loss": 2.7934,
|
| 2226 |
+
"theoretical_loss": 3.653986572492247,
|
| 2227 |
+
"tokens_seen": 986316800
|
| 2228 |
+
},
|
| 2229 |
+
{
|
| 2230 |
+
"epoch": 0.35,
|
| 2231 |
+
"learning_rate": 0.00045721108379042936,
|
| 2232 |
+
"loss": 2.7583,
|
| 2233 |
+
"theoretical_loss": 3.65283080741328,
|
| 2234 |
+
"tokens_seen": 989593600
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 0.35,
|
| 2238 |
+
"learning_rate": 0.00045640013240022694,
|
| 2239 |
+
"loss": 2.7744,
|
| 2240 |
+
"theoretical_loss": 3.6516799305893866,
|
| 2241 |
+
"tokens_seen": 992870400
|
| 2242 |
+
},
|
| 2243 |
+
{
|
| 2244 |
+
"epoch": 0.36,
|
| 2245 |
+
"learning_rate": 0.00045557263098165306,
|
| 2246 |
+
"loss": 2.7683,
|
| 2247 |
+
"theoretical_loss": 3.6505339053151076,
|
| 2248 |
+
"tokens_seen": 996147200
|
| 2249 |
+
},
|
| 2250 |
+
{
|
| 2251 |
+
"epoch": 0.36,
|
| 2252 |
+
"learning_rate": 0.0004547451295630793,
|
| 2253 |
+
"loss": 2.7755,
|
| 2254 |
+
"theoretical_loss": 3.649392695280186,
|
| 2255 |
+
"tokens_seen": 999424000
|
| 2256 |
+
},
|
| 2257 |
+
{
|
| 2258 |
+
"epoch": 0.36,
|
| 2259 |
+
"learning_rate": 0.0004539176281445054,
|
| 2260 |
+
"loss": 2.7407,
|
| 2261 |
+
"theoretical_loss": 3.6482562645640337,
|
| 2262 |
+
"tokens_seen": 1002700800
|
| 2263 |
+
},
|
| 2264 |
+
{
|
| 2265 |
+
"epoch": 0.36,
|
| 2266 |
+
"learning_rate": 0.0004530901267259315,
|
| 2267 |
+
"loss": 2.7456,
|
| 2268 |
+
"theoretical_loss": 3.6471245776302883,
|
| 2269 |
+
"tokens_seen": 1005977600
|
| 2270 |
+
},
|
| 2271 |
+
{
|
| 2272 |
+
"epoch": 0.36,
|
| 2273 |
+
"learning_rate": 0.00045226262530735764,
|
| 2274 |
+
"loss": 2.7827,
|
| 2275 |
+
"theoretical_loss": 3.6459975993214724,
|
| 2276 |
+
"tokens_seen": 1009254400
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 0.36,
|
| 2280 |
+
"learning_rate": 0.0004514351238887838,
|
| 2281 |
+
"loss": 2.7414,
|
| 2282 |
+
"theoretical_loss": 3.6448752948537377,
|
| 2283 |
+
"tokens_seen": 1012531200
|
| 2284 |
+
},
|
| 2285 |
+
{
|
| 2286 |
+
"epoch": 0.36,
|
| 2287 |
+
"learning_rate": 0.00045060762247020993,
|
| 2288 |
+
"loss": 2.7774,
|
| 2289 |
+
"theoretical_loss": 3.6437576298116996,
|
| 2290 |
+
"tokens_seen": 1015808000
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 0.36,
|
| 2294 |
+
"learning_rate": 0.00044978012105163605,
|
| 2295 |
+
"loss": 2.7705,
|
| 2296 |
+
"theoretical_loss": 3.6426445701433607,
|
| 2297 |
+
"tokens_seen": 1019084800
|
| 2298 |
+
},
|
| 2299 |
+
{
|
| 2300 |
+
"epoch": 0.37,
|
| 2301 |
+
"learning_rate": 0.0004489526196330623,
|
| 2302 |
+
"loss": 2.79,
|
| 2303 |
+
"theoretical_loss": 3.6415360821551226,
|
| 2304 |
+
"tokens_seen": 1022361600
|
| 2305 |
+
},
|
| 2306 |
+
{
|
| 2307 |
+
"epoch": 0.37,
|
| 2308 |
+
"learning_rate": 0.0004481251182144884,
|
| 2309 |
+
"loss": 2.7458,
|
| 2310 |
+
"theoretical_loss": 3.6404321325068754,
|
| 2311 |
+
"tokens_seen": 1025638400
|
| 2312 |
+
},
|
| 2313 |
+
{
|
| 2314 |
+
"epoch": 0.37,
|
| 2315 |
+
"learning_rate": 0.0004472976167959145,
|
| 2316 |
+
"loss": 2.7668,
|
| 2317 |
+
"theoretical_loss": 3.639332688207178,
|
| 2318 |
+
"tokens_seen": 1028915200
|
| 2319 |
+
},
|
| 2320 |
+
{
|
| 2321 |
+
"epoch": 0.37,
|
| 2322 |
+
"learning_rate": 0.00044647011537734063,
|
| 2323 |
+
"loss": 2.7474,
|
| 2324 |
+
"theoretical_loss": 3.6382377166085096,
|
| 2325 |
+
"tokens_seen": 1032192000
|
| 2326 |
+
},
|
| 2327 |
+
{
|
| 2328 |
+
"epoch": 0.37,
|
| 2329 |
+
"learning_rate": 0.00044564261395876675,
|
| 2330 |
+
"loss": 2.7098,
|
| 2331 |
+
"theoretical_loss": 3.6371471854026147,
|
| 2332 |
+
"tokens_seen": 1035468800
|
| 2333 |
+
},
|
| 2334 |
+
{
|
| 2335 |
+
"epoch": 0.37,
|
| 2336 |
+
"learning_rate": 0.00044481511254019287,
|
| 2337 |
+
"loss": 2.7161,
|
| 2338 |
+
"theoretical_loss": 3.6360610626159087,
|
| 2339 |
+
"tokens_seen": 1038745600
|
| 2340 |
+
},
|
| 2341 |
+
{
|
| 2342 |
+
"epoch": 0.37,
|
| 2343 |
+
"learning_rate": 0.0004439876111216191,
|
| 2344 |
+
"loss": 2.6831,
|
| 2345 |
+
"theoretical_loss": 3.634979316604973,
|
| 2346 |
+
"tokens_seen": 1042022400
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"epoch": 0.37,
|
| 2350 |
+
"learning_rate": 0.0004431601097030452,
|
| 2351 |
+
"loss": 2.6983,
|
| 2352 |
+
"theoretical_loss": 3.6339019160521198,
|
| 2353 |
+
"tokens_seen": 1045299200
|
| 2354 |
+
},
|
| 2355 |
+
{
|
| 2356 |
+
"epoch": 0.37,
|
| 2357 |
+
"learning_rate": 0.00044233260828447133,
|
| 2358 |
+
"loss": 2.6901,
|
| 2359 |
+
"theoretical_loss": 3.632828829961029,
|
| 2360 |
+
"tokens_seen": 1048576000
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 0.38,
|
| 2364 |
+
"learning_rate": 0.00044150510686589745,
|
| 2365 |
+
"loss": 2.7065,
|
| 2366 |
+
"theoretical_loss": 3.631760027652461,
|
| 2367 |
+
"tokens_seen": 1051852800
|
| 2368 |
+
},
|
| 2369 |
+
{
|
| 2370 |
+
"epoch": 0.38,
|
| 2371 |
+
"learning_rate": 0.0004406776054473236,
|
| 2372 |
+
"loss": 2.7373,
|
| 2373 |
+
"theoretical_loss": 3.630695478760034,
|
| 2374 |
+
"tokens_seen": 1055129600
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"epoch": 0.38,
|
| 2378 |
+
"learning_rate": 0.00043985010402874974,
|
| 2379 |
+
"loss": 2.709,
|
| 2380 |
+
"theoretical_loss": 3.6296351532260767,
|
| 2381 |
+
"tokens_seen": 1058406400
|
| 2382 |
+
},
|
| 2383 |
+
{
|
| 2384 |
+
"epoch": 0.38,
|
| 2385 |
+
"learning_rate": 0.0004390226026101759,
|
| 2386 |
+
"loss": 2.7033,
|
| 2387 |
+
"theoretical_loss": 3.6285790212975435,
|
| 2388 |
+
"tokens_seen": 1061683200
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"epoch": 0.38,
|
| 2392 |
+
"learning_rate": 0.0004381951011916021,
|
| 2393 |
+
"loss": 2.6803,
|
| 2394 |
+
"theoretical_loss": 3.6275270535220008,
|
| 2395 |
+
"tokens_seen": 1064960000
|
| 2396 |
+
},
|
| 2397 |
+
{
|
| 2398 |
+
"epoch": 0.38,
|
| 2399 |
+
"learning_rate": 0.0004373675997730282,
|
| 2400 |
+
"loss": 2.7044,
|
| 2401 |
+
"theoretical_loss": 3.626479220743673,
|
| 2402 |
+
"tokens_seen": 1068236800
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 0.38,
|
| 2406 |
+
"learning_rate": 0.0004365400983544543,
|
| 2407 |
+
"loss": 2.7078,
|
| 2408 |
+
"theoretical_loss": 3.6254354940995586,
|
| 2409 |
+
"tokens_seen": 1071513600
|
| 2410 |
+
},
|
| 2411 |
+
{
|
| 2412 |
+
"epoch": 0.38,
|
| 2413 |
+
"learning_rate": 0.00043571259693588044,
|
| 2414 |
+
"loss": 2.7014,
|
| 2415 |
+
"theoretical_loss": 3.624395845015602,
|
| 2416 |
+
"tokens_seen": 1074790400
|
| 2417 |
+
},
|
| 2418 |
+
{
|
| 2419 |
+
"epoch": 0.39,
|
| 2420 |
+
"learning_rate": 0.00043488509551730656,
|
| 2421 |
+
"loss": 2.6599,
|
| 2422 |
+
"theoretical_loss": 3.6233602452029348,
|
| 2423 |
+
"tokens_seen": 1078067200
|
| 2424 |
+
},
|
| 2425 |
+
{
|
| 2426 |
+
"epoch": 0.39,
|
| 2427 |
+
"learning_rate": 0.0004340575940987327,
|
| 2428 |
+
"loss": 2.6797,
|
| 2429 |
+
"theoretical_loss": 3.6223286666541683,
|
| 2430 |
+
"tokens_seen": 1081344000
|
| 2431 |
+
},
|
| 2432 |
+
{
|
| 2433 |
+
"epoch": 0.39,
|
| 2434 |
+
"learning_rate": 0.0004332300926801589,
|
| 2435 |
+
"loss": 2.7048,
|
| 2436 |
+
"theoretical_loss": 3.621301081639753,
|
| 2437 |
+
"tokens_seen": 1084620800
|
| 2438 |
+
},
|
| 2439 |
+
{
|
| 2440 |
+
"epoch": 0.39,
|
| 2441 |
+
"learning_rate": 0.000432402591261585,
|
| 2442 |
+
"loss": 2.7439,
|
| 2443 |
+
"theoretical_loss": 3.6202774627043923,
|
| 2444 |
+
"tokens_seen": 1087897600
|
| 2445 |
+
},
|
| 2446 |
+
{
|
| 2447 |
+
"epoch": 0.39,
|
| 2448 |
+
"learning_rate": 0.00043157508984301114,
|
| 2449 |
+
"loss": 2.6985,
|
| 2450 |
+
"theoretical_loss": 3.619257782663513,
|
| 2451 |
+
"tokens_seen": 1091174400
|
| 2452 |
+
},
|
| 2453 |
+
{
|
| 2454 |
+
"epoch": 0.39,
|
| 2455 |
+
"learning_rate": 0.0004307475884244373,
|
| 2456 |
+
"loss": 2.7095,
|
| 2457 |
+
"theoretical_loss": 3.618242014599793,
|
| 2458 |
+
"tokens_seen": 1094451200
|
| 2459 |
+
},
|
| 2460 |
+
{
|
| 2461 |
+
"epoch": 0.39,
|
| 2462 |
+
"learning_rate": 0.00042992008700586343,
|
| 2463 |
+
"loss": 2.6983,
|
| 2464 |
+
"theoretical_loss": 3.617230131859743,
|
| 2465 |
+
"tokens_seen": 1097728000
|
| 2466 |
+
},
|
| 2467 |
+
{
|
| 2468 |
+
"epoch": 0.39,
|
| 2469 |
+
"learning_rate": 0.00042909258558728955,
|
| 2470 |
+
"loss": 2.6823,
|
| 2471 |
+
"theoretical_loss": 3.6162221080503416,
|
| 2472 |
+
"tokens_seen": 1101004800
|
| 2473 |
+
},
|
| 2474 |
+
{
|
| 2475 |
+
"epoch": 0.39,
|
| 2476 |
+
"learning_rate": 0.0004282650841687157,
|
| 2477 |
+
"loss": 2.6665,
|
| 2478 |
+
"theoretical_loss": 3.615217917035726,
|
| 2479 |
+
"tokens_seen": 1104281600
|
| 2480 |
+
},
|
| 2481 |
+
{
|
| 2482 |
+
"epoch": 0.4,
|
| 2483 |
+
"learning_rate": 0.0004274375827501419,
|
| 2484 |
+
"loss": 2.6981,
|
| 2485 |
+
"theoretical_loss": 3.614217532933929,
|
| 2486 |
+
"tokens_seen": 1107558400
|
| 2487 |
+
},
|
| 2488 |
+
{
|
| 2489 |
+
"epoch": 0.4,
|
| 2490 |
+
"learning_rate": 0.000426610081331568,
|
| 2491 |
+
"loss": 2.6861,
|
| 2492 |
+
"theoretical_loss": 3.6132209301136715,
|
| 2493 |
+
"tokens_seen": 1110835200
|
| 2494 |
+
},
|
| 2495 |
+
{
|
| 2496 |
+
"epoch": 0.4,
|
| 2497 |
+
"learning_rate": 0.00042578257991299413,
|
| 2498 |
+
"loss": 2.6717,
|
| 2499 |
+
"theoretical_loss": 3.612228083191205,
|
| 2500 |
+
"tokens_seen": 1114112000
|
| 2501 |
+
},
|
| 2502 |
+
{
|
| 2503 |
+
"epoch": 0.4,
|
| 2504 |
+
"learning_rate": 0.00042495507849442025,
|
| 2505 |
+
"loss": 2.7291,
|
| 2506 |
+
"theoretical_loss": 3.611238967027199,
|
| 2507 |
+
"tokens_seen": 1117388800
|
| 2508 |
+
},
|
| 2509 |
+
{
|
| 2510 |
+
"epoch": 0.4,
|
| 2511 |
+
"learning_rate": 0.00042412757707584636,
|
| 2512 |
+
"loss": 2.7692,
|
| 2513 |
+
"theoretical_loss": 3.610253556723679,
|
| 2514 |
+
"tokens_seen": 1120665600
|
| 2515 |
+
},
|
| 2516 |
+
{
|
| 2517 |
+
"epoch": 0.4,
|
| 2518 |
+
"learning_rate": 0.0004233000756572726,
|
| 2519 |
+
"loss": 2.7318,
|
| 2520 |
+
"theoretical_loss": 3.609271827621014,
|
| 2521 |
+
"tokens_seen": 1123942400
|
| 2522 |
+
},
|
| 2523 |
+
{
|
| 2524 |
+
"epoch": 0.4,
|
| 2525 |
+
"learning_rate": 0.0004224725742386987,
|
| 2526 |
+
"loss": 2.7372,
|
| 2527 |
+
"theoretical_loss": 3.6082937552949463,
|
| 2528 |
+
"tokens_seen": 1127219200
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"epoch": 0.4,
|
| 2532 |
+
"learning_rate": 0.00042164507282012483,
|
| 2533 |
+
"loss": 2.7052,
|
| 2534 |
+
"theoretical_loss": 3.607319315553669,
|
| 2535 |
+
"tokens_seen": 1130496000
|
| 2536 |
+
},
|
| 2537 |
+
{
|
| 2538 |
+
"epoch": 0.4,
|
| 2539 |
+
"learning_rate": 0.00042081757140155095,
|
| 2540 |
+
"loss": 2.668,
|
| 2541 |
+
"theoretical_loss": 3.6063484844349456,
|
| 2542 |
+
"tokens_seen": 1133772800
|
| 2543 |
+
},
|
| 2544 |
+
{
|
| 2545 |
+
"epoch": 0.41,
|
| 2546 |
+
"learning_rate": 0.0004199900699829771,
|
| 2547 |
+
"loss": 2.6911,
|
| 2548 |
+
"theoretical_loss": 3.605381238203279,
|
| 2549 |
+
"tokens_seen": 1137049600
|
| 2550 |
+
},
|
| 2551 |
+
{
|
| 2552 |
+
"epoch": 0.41,
|
| 2553 |
+
"learning_rate": 0.00041916256856440324,
|
| 2554 |
+
"loss": 2.7014,
|
| 2555 |
+
"theoretical_loss": 3.604417553347117,
|
| 2556 |
+
"tokens_seen": 1140326400
|
| 2557 |
+
},
|
| 2558 |
+
{
|
| 2559 |
+
"epoch": 0.41,
|
| 2560 |
+
"learning_rate": 0.0004183350671458294,
|
| 2561 |
+
"loss": 2.6926,
|
| 2562 |
+
"theoretical_loss": 3.603457406576106,
|
| 2563 |
+
"tokens_seen": 1143603200
|
| 2564 |
+
},
|
| 2565 |
+
{
|
| 2566 |
+
"epoch": 0.41,
|
| 2567 |
+
"objective/train/avg_token_score": 0.022877871990203857,
|
| 2568 |
+
"objective/train/avg_weight": 0.97539883852005,
|
| 2569 |
+
"objective/train/docs_used": 649861,
|
| 2570 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 2571 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 2572 |
+
"objective/train/original_loss": 2.6508138179779053,
|
| 2573 |
+
"objective/train/std_weight": 0.10265526175498962,
|
| 2574 |
+
"objective/train/theoretical_loss": 3.602500774818379,
|
| 2575 |
+
"objective/train/tokens_used": 1167340000,
|
| 2576 |
+
"theoretical_loss": 3.602500774818379,
|
| 2577 |
+
"tokens_seen": 1146880000
|
| 2578 |
+
},
|
| 2579 |
+
{
|
| 2580 |
+
"epoch": 0.41,
|
| 2581 |
+
"learning_rate": 0.00041750756572725553,
|
| 2582 |
+
"loss": 2.6792,
|
| 2583 |
+
"theoretical_loss": 3.602500774818379,
|
| 2584 |
+
"tokens_seen": 1146880000
|
| 2585 |
+
},
|
| 2586 |
+
{
|
| 2587 |
+
"epoch": 0.41,
|
| 2588 |
+
"learning_rate": 0.0004166800643086817,
|
| 2589 |
+
"loss": 2.7197,
|
| 2590 |
+
"theoretical_loss": 3.601547635217892,
|
| 2591 |
+
"tokens_seen": 1150156800
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 0.41,
|
| 2595 |
+
"learning_rate": 0.0004158525628901078,
|
| 2596 |
+
"loss": 2.7201,
|
| 2597 |
+
"theoretical_loss": 3.6005979651317976,
|
| 2598 |
+
"tokens_seen": 1153433600
|
| 2599 |
+
},
|
| 2600 |
+
{
|
| 2601 |
+
"epoch": 0.41,
|
| 2602 |
+
"learning_rate": 0.00041502506147153394,
|
| 2603 |
+
"loss": 2.7286,
|
| 2604 |
+
"theoretical_loss": 3.599651742127855,
|
| 2605 |
+
"tokens_seen": 1156710400
|
| 2606 |
+
},
|
| 2607 |
+
{
|
| 2608 |
+
"epoch": 0.41,
|
| 2609 |
+
"learning_rate": 0.00041419756005296005,
|
| 2610 |
+
"loss": 2.681,
|
| 2611 |
+
"theoretical_loss": 3.5987089439818805,
|
| 2612 |
+
"tokens_seen": 1159987200
|
| 2613 |
+
},
|
| 2614 |
+
{
|
| 2615 |
+
"epoch": 0.42,
|
| 2616 |
+
"learning_rate": 0.00041337005863438617,
|
| 2617 |
+
"loss": 2.6621,
|
| 2618 |
+
"theoretical_loss": 3.5977695486752426,
|
| 2619 |
+
"tokens_seen": 1163264000
|
| 2620 |
+
},
|
| 2621 |
+
{
|
| 2622 |
+
"epoch": 0.42,
|
| 2623 |
+
"learning_rate": 0.0004125425572158124,
|
| 2624 |
+
"loss": 2.6926,
|
| 2625 |
+
"theoretical_loss": 3.596833534392379,
|
| 2626 |
+
"tokens_seen": 1166540800
|
| 2627 |
+
},
|
| 2628 |
+
{
|
| 2629 |
+
"epoch": 0.42,
|
| 2630 |
+
"learning_rate": 0.0004117150557972385,
|
| 2631 |
+
"loss": 2.7096,
|
| 2632 |
+
"theoretical_loss": 3.595900879518368,
|
| 2633 |
+
"tokens_seen": 1169817600
|
| 2634 |
+
},
|
| 2635 |
+
{
|
| 2636 |
+
"epoch": 0.42,
|
| 2637 |
+
"learning_rate": 0.00041088755437866464,
|
| 2638 |
+
"loss": 2.7241,
|
| 2639 |
+
"theoretical_loss": 3.594971562636521,
|
| 2640 |
+
"tokens_seen": 1173094400
|
| 2641 |
+
},
|
| 2642 |
+
{
|
| 2643 |
+
"epoch": 0.42,
|
| 2644 |
+
"learning_rate": 0.00041006005296009075,
|
| 2645 |
+
"loss": 2.6816,
|
| 2646 |
+
"theoretical_loss": 3.5940455625260226,
|
| 2647 |
+
"tokens_seen": 1176371200
|
| 2648 |
+
},
|
| 2649 |
+
{
|
| 2650 |
+
"epoch": 0.42,
|
| 2651 |
+
"learning_rate": 0.0004092325515415169,
|
| 2652 |
+
"loss": 2.6928,
|
| 2653 |
+
"theoretical_loss": 3.5931228581595938,
|
| 2654 |
+
"tokens_seen": 1179648000
|
| 2655 |
+
},
|
| 2656 |
+
{
|
| 2657 |
+
"epoch": 0.42,
|
| 2658 |
+
"learning_rate": 0.00040840505012294305,
|
| 2659 |
+
"loss": 2.6899,
|
| 2660 |
+
"theoretical_loss": 3.5922034287011995,
|
| 2661 |
+
"tokens_seen": 1182924800
|
| 2662 |
+
},
|
| 2663 |
+
{
|
| 2664 |
+
"epoch": 0.42,
|
| 2665 |
+
"learning_rate": 0.0004075775487043692,
|
| 2666 |
+
"loss": 2.6567,
|
| 2667 |
+
"theoretical_loss": 3.5912872535037828,
|
| 2668 |
+
"tokens_seen": 1186201600
|
| 2669 |
+
},
|
| 2670 |
+
{
|
| 2671 |
+
"epoch": 0.42,
|
| 2672 |
+
"learning_rate": 0.0004067500472857954,
|
| 2673 |
+
"loss": 2.739,
|
| 2674 |
+
"theoretical_loss": 3.590374312107035,
|
| 2675 |
+
"tokens_seen": 1189478400
|
| 2676 |
+
},
|
| 2677 |
+
{
|
| 2678 |
+
"epoch": 0.43,
|
| 2679 |
+
"learning_rate": 0.0004059225458672215,
|
| 2680 |
+
"loss": 2.724,
|
| 2681 |
+
"theoretical_loss": 3.5894645842351993,
|
| 2682 |
+
"tokens_seen": 1192755200
|
| 2683 |
+
},
|
| 2684 |
+
{
|
| 2685 |
+
"epoch": 0.43,
|
| 2686 |
+
"learning_rate": 0.0004050950444486476,
|
| 2687 |
+
"loss": 2.7097,
|
| 2688 |
+
"theoretical_loss": 3.588558049794902,
|
| 2689 |
+
"tokens_seen": 1196032000
|
| 2690 |
+
},
|
| 2691 |
+
{
|
| 2692 |
+
"epoch": 0.43,
|
| 2693 |
+
"learning_rate": 0.00040426754303007374,
|
| 2694 |
+
"loss": 2.689,
|
| 2695 |
+
"theoretical_loss": 3.5876546888730187,
|
| 2696 |
+
"tokens_seen": 1199308800
|
| 2697 |
+
},
|
| 2698 |
+
{
|
| 2699 |
+
"epoch": 0.43,
|
| 2700 |
+
"learning_rate": 0.00040344004161149986,
|
| 2701 |
+
"loss": 2.7083,
|
| 2702 |
+
"theoretical_loss": 3.5867544817345713,
|
| 2703 |
+
"tokens_seen": 1202585600
|
| 2704 |
+
},
|
| 2705 |
+
{
|
| 2706 |
+
"epoch": 0.43,
|
| 2707 |
+
"learning_rate": 0.0004026125401929261,
|
| 2708 |
+
"loss": 2.7336,
|
| 2709 |
+
"theoretical_loss": 3.585857408820652,
|
| 2710 |
+
"tokens_seen": 1205862400
|
| 2711 |
+
},
|
| 2712 |
+
{
|
| 2713 |
+
"epoch": 0.43,
|
| 2714 |
+
"learning_rate": 0.00040180158880272367,
|
| 2715 |
+
"loss": 2.7425,
|
| 2716 |
+
"theoretical_loss": 3.58496345074638,
|
| 2717 |
+
"tokens_seen": 1209139200
|
| 2718 |
+
},
|
| 2719 |
+
{
|
| 2720 |
+
"epoch": 0.43,
|
| 2721 |
+
"learning_rate": 0.0004009740873841498,
|
| 2722 |
+
"loss": 2.7536,
|
| 2723 |
+
"theoretical_loss": 3.5840725882988873,
|
| 2724 |
+
"tokens_seen": 1212416000
|
| 2725 |
+
},
|
| 2726 |
+
{
|
| 2727 |
+
"epoch": 0.43,
|
| 2728 |
+
"learning_rate": 0.00040014658596557596,
|
| 2729 |
+
"loss": 2.7189,
|
| 2730 |
+
"theoretical_loss": 3.5831848024353317,
|
| 2731 |
+
"tokens_seen": 1215692800
|
| 2732 |
+
},
|
| 2733 |
+
{
|
| 2734 |
+
"epoch": 0.44,
|
| 2735 |
+
"learning_rate": 0.0003993190845470021,
|
| 2736 |
+
"loss": 2.7442,
|
| 2737 |
+
"theoretical_loss": 3.5823000742809374,
|
| 2738 |
+
"tokens_seen": 1218969600
|
| 2739 |
+
},
|
| 2740 |
+
{
|
| 2741 |
+
"epoch": 0.44,
|
| 2742 |
+
"learning_rate": 0.0003984915831284282,
|
| 2743 |
+
"loss": 2.7158,
|
| 2744 |
+
"theoretical_loss": 3.5814183851270673,
|
| 2745 |
+
"tokens_seen": 1222246400
|
| 2746 |
+
},
|
| 2747 |
+
{
|
| 2748 |
+
"epoch": 0.44,
|
| 2749 |
+
"learning_rate": 0.0003976640817098543,
|
| 2750 |
+
"loss": 2.7064,
|
| 2751 |
+
"theoretical_loss": 3.5805397164293167,
|
| 2752 |
+
"tokens_seen": 1225523200
|
| 2753 |
+
},
|
| 2754 |
+
{
|
| 2755 |
+
"epoch": 0.44,
|
| 2756 |
+
"learning_rate": 0.00039683658029128054,
|
| 2757 |
+
"loss": 2.6876,
|
| 2758 |
+
"theoretical_loss": 3.5796640498056407,
|
| 2759 |
+
"tokens_seen": 1228800000
|
| 2760 |
+
},
|
| 2761 |
+
{
|
| 2762 |
+
"epoch": 0.44,
|
| 2763 |
+
"learning_rate": 0.00039600907887270666,
|
| 2764 |
+
"loss": 2.696,
|
| 2765 |
+
"theoretical_loss": 3.5787913670345013,
|
| 2766 |
+
"tokens_seen": 1232076800
|
| 2767 |
+
},
|
| 2768 |
+
{
|
| 2769 |
+
"epoch": 0.44,
|
| 2770 |
+
"learning_rate": 0.00039519812748250424,
|
| 2771 |
+
"loss": 2.6741,
|
| 2772 |
+
"theoretical_loss": 3.577921650053045,
|
| 2773 |
+
"tokens_seen": 1235353600
|
| 2774 |
+
},
|
| 2775 |
+
{
|
| 2776 |
+
"epoch": 0.44,
|
| 2777 |
+
"learning_rate": 0.00039437062606393036,
|
| 2778 |
+
"loss": 2.6706,
|
| 2779 |
+
"theoretical_loss": 3.577054880955303,
|
| 2780 |
+
"tokens_seen": 1238630400
|
| 2781 |
+
},
|
| 2782 |
+
{
|
| 2783 |
+
"epoch": 0.44,
|
| 2784 |
+
"learning_rate": 0.00039354312464535653,
|
| 2785 |
+
"loss": 2.657,
|
| 2786 |
+
"theoretical_loss": 3.5761910419904193,
|
| 2787 |
+
"tokens_seen": 1241907200
|
| 2788 |
+
},
|
| 2789 |
+
{
|
| 2790 |
+
"epoch": 0.44,
|
| 2791 |
+
"learning_rate": 0.00039271562322678265,
|
| 2792 |
+
"loss": 2.7193,
|
| 2793 |
+
"theoretical_loss": 3.5753301155609014,
|
| 2794 |
+
"tokens_seen": 1245184000
|
| 2795 |
+
},
|
| 2796 |
+
{
|
| 2797 |
+
"epoch": 0.45,
|
| 2798 |
+
"learning_rate": 0.0003918881218082088,
|
| 2799 |
+
"loss": 2.6606,
|
| 2800 |
+
"theoretical_loss": 3.574472084220896,
|
| 2801 |
+
"tokens_seen": 1248460800
|
| 2802 |
+
},
|
| 2803 |
+
{
|
| 2804 |
+
"epoch": 0.45,
|
| 2805 |
+
"learning_rate": 0.00039106062038963494,
|
| 2806 |
+
"loss": 2.6497,
|
| 2807 |
+
"theoretical_loss": 3.5736169306744885,
|
| 2808 |
+
"tokens_seen": 1251737600
|
| 2809 |
+
},
|
| 2810 |
+
{
|
| 2811 |
+
"epoch": 0.45,
|
| 2812 |
+
"learning_rate": 0.0003902331189710611,
|
| 2813 |
+
"loss": 2.6164,
|
| 2814 |
+
"theoretical_loss": 3.572764637774024,
|
| 2815 |
+
"tokens_seen": 1255014400
|
| 2816 |
+
},
|
| 2817 |
+
{
|
| 2818 |
+
"epoch": 0.45,
|
| 2819 |
+
"learning_rate": 0.00038940561755248723,
|
| 2820 |
+
"loss": 2.6288,
|
| 2821 |
+
"theoretical_loss": 3.571915188518457,
|
| 2822 |
+
"tokens_seen": 1258291200
|
| 2823 |
+
},
|
| 2824 |
+
{
|
| 2825 |
+
"epoch": 0.45,
|
| 2826 |
+
"learning_rate": 0.00038857811613391335,
|
| 2827 |
+
"loss": 2.6475,
|
| 2828 |
+
"theoretical_loss": 3.571068566051716,
|
| 2829 |
+
"tokens_seen": 1261568000
|
| 2830 |
+
},
|
| 2831 |
+
{
|
| 2832 |
+
"epoch": 0.45,
|
| 2833 |
+
"learning_rate": 0.00038775061471533947,
|
| 2834 |
+
"loss": 2.6866,
|
| 2835 |
+
"theoretical_loss": 3.5702247536610976,
|
| 2836 |
+
"tokens_seen": 1264844800
|
| 2837 |
+
},
|
| 2838 |
+
{
|
| 2839 |
+
"epoch": 0.45,
|
| 2840 |
+
"learning_rate": 0.0003869231132967657,
|
| 2841 |
+
"loss": 2.667,
|
| 2842 |
+
"theoretical_loss": 3.5693837347756783,
|
| 2843 |
+
"tokens_seen": 1268121600
|
| 2844 |
+
},
|
| 2845 |
+
{
|
| 2846 |
+
"epoch": 0.45,
|
| 2847 |
+
"learning_rate": 0.0003860956118781918,
|
| 2848 |
+
"loss": 2.6401,
|
| 2849 |
+
"theoretical_loss": 3.5685454929647475,
|
| 2850 |
+
"tokens_seen": 1271398400
|
| 2851 |
+
},
|
| 2852 |
+
{
|
| 2853 |
+
"epoch": 0.46,
|
| 2854 |
+
"learning_rate": 0.00038526811045961793,
|
| 2855 |
+
"loss": 2.6509,
|
| 2856 |
+
"theoretical_loss": 3.5677100119362675,
|
| 2857 |
+
"tokens_seen": 1274675200
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 0.46,
|
| 2861 |
+
"learning_rate": 0.00038444060904104405,
|
| 2862 |
+
"loss": 2.6335,
|
| 2863 |
+
"theoretical_loss": 3.566877275535345,
|
| 2864 |
+
"tokens_seen": 1277952000
|
| 2865 |
+
},
|
| 2866 |
+
{
|
| 2867 |
+
"epoch": 0.46,
|
| 2868 |
+
"learning_rate": 0.00038361310762247017,
|
| 2869 |
+
"loss": 2.6285,
|
| 2870 |
+
"theoretical_loss": 3.566047267742733,
|
| 2871 |
+
"tokens_seen": 1281228800
|
| 2872 |
+
},
|
| 2873 |
+
{
|
| 2874 |
+
"epoch": 0.46,
|
| 2875 |
+
"learning_rate": 0.00038278560620389634,
|
| 2876 |
+
"loss": 2.6087,
|
| 2877 |
+
"theoretical_loss": 3.5652199726733453,
|
| 2878 |
+
"tokens_seen": 1284505600
|
| 2879 |
+
},
|
| 2880 |
+
{
|
| 2881 |
+
"epoch": 0.46,
|
| 2882 |
+
"learning_rate": 0.0003819581047853225,
|
| 2883 |
+
"loss": 2.6032,
|
| 2884 |
+
"theoretical_loss": 3.564395374574796,
|
| 2885 |
+
"tokens_seen": 1287782400
|
| 2886 |
+
},
|
| 2887 |
+
{
|
| 2888 |
+
"epoch": 0.46,
|
| 2889 |
+
"learning_rate": 0.00038113060336674863,
|
| 2890 |
+
"loss": 2.6624,
|
| 2891 |
+
"theoretical_loss": 3.5635734578259557,
|
| 2892 |
+
"tokens_seen": 1291059200
|
| 2893 |
+
},
|
| 2894 |
+
{
|
| 2895 |
+
"epoch": 0.46,
|
| 2896 |
+
"learning_rate": 0.0003803031019481748,
|
| 2897 |
+
"loss": 2.6763,
|
| 2898 |
+
"theoretical_loss": 3.5627542069355282,
|
| 2899 |
+
"tokens_seen": 1294336000
|
| 2900 |
+
},
|
| 2901 |
+
{
|
| 2902 |
+
"epoch": 0.46,
|
| 2903 |
+
"learning_rate": 0.0003794756005296009,
|
| 2904 |
+
"loss": 2.6398,
|
| 2905 |
+
"theoretical_loss": 3.5619376065406474,
|
| 2906 |
+
"tokens_seen": 1297612800
|
| 2907 |
+
},
|
| 2908 |
+
{
|
| 2909 |
+
"epoch": 0.46,
|
| 2910 |
+
"learning_rate": 0.00037864809911102704,
|
| 2911 |
+
"loss": 2.6559,
|
| 2912 |
+
"theoretical_loss": 3.5611236414054868,
|
| 2913 |
+
"tokens_seen": 1300889600
|
| 2914 |
+
},
|
| 2915 |
+
{
|
| 2916 |
+
"epoch": 0.47,
|
| 2917 |
+
"learning_rate": 0.00037782059769245316,
|
| 2918 |
+
"loss": 2.6598,
|
| 2919 |
+
"theoretical_loss": 3.560312296419899,
|
| 2920 |
+
"tokens_seen": 1304166400
|
| 2921 |
+
},
|
| 2922 |
+
{
|
| 2923 |
+
"epoch": 0.47,
|
| 2924 |
+
"learning_rate": 0.0003769930962738793,
|
| 2925 |
+
"loss": 2.6584,
|
| 2926 |
+
"theoretical_loss": 3.55950355659806,
|
| 2927 |
+
"tokens_seen": 1307443200
|
| 2928 |
+
},
|
| 2929 |
+
{
|
| 2930 |
+
"debugging/Self-BLEU-5": 0.49020908264157476,
|
| 2931 |
+
"debugging/distinct-1-grams": 0.768901113497886,
|
| 2932 |
+
"debugging/distinct-2-grams": 0.9428782333551957,
|
| 2933 |
+
"debugging/entropy-1-grams": 6.085999550681761,
|
| 2934 |
+
"debugging/entropy-2-grams": 7.0033060167714964,
|
| 2935 |
+
"debugging/length": 490.2352941176471,
|
| 2936 |
+
"debugging/num_segments": 17,
|
| 2937 |
+
"epoch": 0.47,
|
| 2938 |
+
"objective/train/avg_token_score": 0.02056093141436577,
|
| 2939 |
+
"objective/train/avg_weight": 0.977873682975769,
|
| 2940 |
+
"objective/train/docs_used": 741674,
|
| 2941 |
+
"objective/train/instantaneous_batch_size": 32,
|
| 2942 |
+
"objective/train/instantaneous_microbatch_size": 32768,
|
| 2943 |
+
"objective/train/original_loss": 3.746752977371216,
|
| 2944 |
+
"objective/train/std_weight": 0.11813095211982727,
|
| 2945 |
+
"objective/train/theoretical_loss": 3.558697407077142,
|
| 2946 |
+
"objective/train/tokens_used": 1331180000,
|
| 2947 |
+
"theoretical_loss": 3.558697407077142,
|
| 2948 |
+
"tokens_seen": 1310720000
|
| 2949 |
+
},
|
| 2950 |
+
{
|
| 2951 |
+
"epoch": 0.47,
|
| 2952 |
+
"learning_rate": 0.0003761655948553055,
|
| 2953 |
+
"loss": 2.6483,
|
| 2954 |
+
"theoretical_loss": 3.558697407077142,
|
| 2955 |
+
"tokens_seen": 1310720000
|
| 2956 |
+
},
|
| 2957 |
+
{
|
| 2958 |
+
"epoch": 0.47,
|
| 2959 |
+
"learning_rate": 0.0003753380934367316,
|
| 2960 |
+
"loss": 2.6646,
|
| 2961 |
+
"theoretical_loss": 3.5578938331159975,
|
| 2962 |
+
"tokens_seen": 1313996800
|
| 2963 |
+
},
|
| 2964 |
+
{
|
| 2965 |
+
"epoch": 0.47,
|
| 2966 |
+
"learning_rate": 0.00037451059201815774,
|
| 2967 |
+
"loss": 2.6382,
|
| 2968 |
+
"theoretical_loss": 3.557092820093863,
|
| 2969 |
+
"tokens_seen": 1317273600
|
| 2970 |
+
},
|
| 2971 |
+
{
|
| 2972 |
+
"epoch": 0.47,
|
| 2973 |
+
"learning_rate": 0.00037368309059958386,
|
| 2974 |
+
"loss": 2.6577,
|
| 2975 |
+
"theoretical_loss": 3.556294353509079,
|
| 2976 |
+
"tokens_seen": 1320550400
|
| 2977 |
+
},
|
| 2978 |
+
{
|
| 2979 |
+
"epoch": 0.47,
|
| 2980 |
+
"learning_rate": 0.00037285558918101,
|
| 2981 |
+
"loss": 2.6493,
|
| 2982 |
+
"theoretical_loss": 3.555498418977828,
|
| 2983 |
+
"tokens_seen": 1323827200
|
| 2984 |
+
},
|
| 2985 |
+
{
|
| 2986 |
+
"epoch": 0.47,
|
| 2987 |
+
"learning_rate": 0.00037202808776243615,
|
| 2988 |
+
"loss": 2.6434,
|
| 2989 |
+
"theoretical_loss": 3.5547050022328874,
|
| 2990 |
+
"tokens_seen": 1327104000
|
| 2991 |
+
},
|
| 2992 |
+
{
|
| 2993 |
+
"epoch": 0.48,
|
| 2994 |
+
"learning_rate": 0.0003712005863438623,
|
| 2995 |
+
"loss": 2.639,
|
| 2996 |
+
"theoretical_loss": 3.553914089122399,
|
| 2997 |
+
"tokens_seen": 1330380800
|
| 2998 |
+
},
|
| 2999 |
+
{
|
| 3000 |
+
"epoch": 0.48,
|
| 3001 |
+
"learning_rate": 0.00037037308492528844,
|
| 3002 |
+
"loss": 2.6797,
|
| 3003 |
+
"theoretical_loss": 3.553125665608655,
|
| 3004 |
+
"tokens_seen": 1333657600
|
| 3005 |
+
},
|
| 3006 |
+
{
|
| 3007 |
+
"epoch": 0.48,
|
| 3008 |
+
"learning_rate": 0.0003695455835067146,
|
| 3009 |
+
"loss": 2.6617,
|
| 3010 |
+
"theoretical_loss": 3.5523397177669005,
|
| 3011 |
+
"tokens_seen": 1336934400
|
| 3012 |
+
},
|
| 3013 |
+
{
|
| 3014 |
+
"epoch": 0.48,
|
| 3015 |
+
"learning_rate": 0.00036871808208814073,
|
| 3016 |
+
"loss": 2.6747,
|
| 3017 |
+
"theoretical_loss": 3.551556231784149,
|
| 3018 |
+
"tokens_seen": 1340211200
|
| 3019 |
+
},
|
| 3020 |
+
{
|
| 3021 |
+
"epoch": 0.48,
|
| 3022 |
+
"learning_rate": 0.00036789058066956685,
|
| 3023 |
+
"loss": 2.6276,
|
| 3024 |
+
"theoretical_loss": 3.5507751939580148,
|
| 3025 |
+
"tokens_seen": 1343488000
|
| 3026 |
+
},
|
| 3027 |
+
{
|
| 3028 |
+
"epoch": 0.48,
|
| 3029 |
+
"learning_rate": 0.00036706307925099297,
|
| 3030 |
+
"loss": 2.6325,
|
| 3031 |
+
"theoretical_loss": 3.5499965906955606,
|
| 3032 |
+
"tokens_seen": 1346764800
|
| 3033 |
+
},
|
| 3034 |
+
{
|
| 3035 |
+
"epoch": 0.48,
|
| 3036 |
+
"learning_rate": 0.0003662355778324192,
|
| 3037 |
+
"loss": 2.6789,
|
| 3038 |
+
"theoretical_loss": 3.549220408512161,
|
| 3039 |
+
"tokens_seen": 1350041600
|
| 3040 |
+
},
|
| 3041 |
+
{
|
| 3042 |
+
"epoch": 0.48,
|
| 3043 |
+
"learning_rate": 0.0003654080764138453,
|
| 3044 |
+
"loss": 2.6544,
|
| 3045 |
+
"theoretical_loss": 3.5484466340303755,
|
| 3046 |
+
"tokens_seen": 1353318400
|
| 3047 |
+
},
|
| 3048 |
+
{
|
| 3049 |
+
"epoch": 0.48,
|
| 3050 |
+
"learning_rate": 0.00036458057499527143,
|
| 3051 |
+
"loss": 2.6411,
|
| 3052 |
+
"theoretical_loss": 3.547675253978843,
|
| 3053 |
+
"tokens_seen": 1356595200
|
| 3054 |
+
},
|
| 3055 |
+
{
|
| 3056 |
+
"epoch": 0.49,
|
| 3057 |
+
"learning_rate": 0.00036375307357669755,
|
| 3058 |
+
"loss": 2.6751,
|
| 3059 |
+
"theoretical_loss": 3.5469062551911854,
|
| 3060 |
+
"tokens_seen": 1359872000
|
| 3061 |
+
},
|
| 3062 |
+
{
|
| 3063 |
+
"epoch": 0.49,
|
| 3064 |
+
"learning_rate": 0.00036292557215812367,
|
| 3065 |
+
"loss": 2.6842,
|
| 3066 |
+
"theoretical_loss": 3.5461396246049244,
|
| 3067 |
+
"tokens_seen": 1363148800
|
| 3068 |
+
},
|
| 3069 |
+
{
|
| 3070 |
+
"epoch": 0.49,
|
| 3071 |
+
"learning_rate": 0.0003620980707395498,
|
| 3072 |
+
"loss": 2.6504,
|
| 3073 |
+
"theoretical_loss": 3.545375349260419,
|
| 3074 |
+
"tokens_seen": 1366425600
|
| 3075 |
+
},
|
| 3076 |
+
{
|
| 3077 |
+
"epoch": 0.49,
|
| 3078 |
+
"learning_rate": 0.00036127056932097596,
|
| 3079 |
+
"loss": 2.6894,
|
| 3080 |
+
"theoretical_loss": 3.544613416299808,
|
| 3081 |
+
"tokens_seen": 1369702400
|
| 3082 |
+
},
|
| 3083 |
+
{
|
| 3084 |
+
"epoch": 0.49,
|
| 3085 |
+
"learning_rate": 0.00036044306790240213,
|
| 3086 |
+
"loss": 2.6629,
|
| 3087 |
+
"theoretical_loss": 3.5438538129659687,
|
| 3088 |
+
"tokens_seen": 1372979200
|
| 3089 |
+
},
|
| 3090 |
+
{
|
| 3091 |
+
"epoch": 0.49,
|
| 3092 |
+
"learning_rate": 0.00035961556648382825,
|
| 3093 |
+
"loss": 2.6823,
|
| 3094 |
+
"theoretical_loss": 3.5430965266014933,
|
| 3095 |
+
"tokens_seen": 1376256000
|
| 3096 |
+
},
|
| 3097 |
+
{
|
| 3098 |
+
"epoch": 0.49,
|
| 3099 |
+
"learning_rate": 0.0003587880650652544,
|
| 3100 |
+
"loss": 2.6879,
|
| 3101 |
+
"theoretical_loss": 3.5423415446476705,
|
| 3102 |
+
"tokens_seen": 1379532800
|
| 3103 |
+
},
|
| 3104 |
+
{
|
| 3105 |
+
"epoch": 0.49,
|
| 3106 |
+
"learning_rate": 0.00035796056364668054,
|
| 3107 |
+
"loss": 2.6256,
|
| 3108 |
+
"theoretical_loss": 3.541588854643487,
|
| 3109 |
+
"tokens_seen": 1382809600
|
| 3110 |
+
},
|
| 3111 |
+
{
|
| 3112 |
+
"epoch": 0.5,
|
| 3113 |
+
"learning_rate": 0.00035713306222810666,
|
| 3114 |
+
"loss": 2.6521,
|
| 3115 |
+
"theoretical_loss": 3.5408384442246343,
|
| 3116 |
+
"tokens_seen": 1386086400
|
| 3117 |
+
},
|
| 3118 |
+
{
|
| 3119 |
+
"epoch": 0.5,
|
| 3120 |
+
"learning_rate": 0.00035632211083790424,
|
| 3121 |
+
"loss": 2.6065,
|
| 3122 |
+
"theoretical_loss": 3.540090301122535,
|
| 3123 |
+
"tokens_seen": 1389363200
|
| 3124 |
+
},
|
| 3125 |
+
{
|
| 3126 |
+
"epoch": 0.5,
|
| 3127 |
+
"learning_rate": 0.00035549460941933046,
|
| 3128 |
+
"loss": 2.6299,
|
| 3129 |
+
"theoretical_loss": 3.5393444131633762,
|
| 3130 |
+
"tokens_seen": 1392640000
|
| 3131 |
+
},
|
| 3132 |
+
{
|
| 3133 |
+
"epoch": 0.5,
|
| 3134 |
+
"learning_rate": 0.0003546671080007566,
|
| 3135 |
+
"loss": 2.676,
|
| 3136 |
+
"theoretical_loss": 3.5386007682671576,
|
| 3137 |
+
"tokens_seen": 1395916800
|
| 3138 |
+
},
|
| 3139 |
+
{
|
| 3140 |
+
"epoch": 0.5,
|
| 3141 |
+
"learning_rate": 0.0003538396065821827,
|
| 3142 |
+
"loss": 2.6589,
|
| 3143 |
+
"theoretical_loss": 3.5378593544467494,
|
| 3144 |
+
"tokens_seen": 1399193600
|
| 3145 |
+
}
|
| 3146 |
+
],
|
| 3147 |
+
"max_steps": 42724,
|
| 3148 |
+
"num_train_epochs": 9223372036854775807,
|
| 3149 |
+
"total_flos": 7.14460209610752e+17,
|
| 3150 |
+
"trial_name": null,
|
| 3151 |
+
"trial_params": null
|
| 3152 |
+
}
|
checkpoint-21362/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7801351d8e61cb75dc1bbc64f93edd4ad0ce9a48aeb7043b62014bfe9986e726
|
| 3 |
+
size 3451
|
checkpoint-21362/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMAndValueHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": true,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.23.0",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50257
|
| 39 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d7f39fa26de841da6fe89695b4819d6ad5a6ddff087f7d0d2d506360443757a3
|
| 3 |
+
size 510398013
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<|endoftext|>",
|
| 4 |
+
"eos_token": "<|endoftext|>",
|
| 5 |
+
"model_max_length": 1024,
|
| 6 |
+
"name_or_path": "gpt2",
|
| 7 |
+
"special_tokens_map_file": null,
|
| 8 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 9 |
+
"unk_token": "<|endoftext|>"
|
| 10 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7801351d8e61cb75dc1bbc64f93edd4ad0ce9a48aeb7043b62014bfe9986e726
|
| 3 |
+
size 3451
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|