Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- checkpoint-2000/added_tokens.json +3 -0
- checkpoint-2000/chat_template.jinja +47 -0
- checkpoint-2000/config.json +110 -0
- checkpoint-2000/generation_config.json +13 -0
- checkpoint-2000/model-00001-of-00005.safetensors +3 -0
- checkpoint-2000/model-00002-of-00005.safetensors +3 -0
- checkpoint-2000/model-00003-of-00005.safetensors +3 -0
- checkpoint-2000/model-00004-of-00005.safetensors +3 -0
- checkpoint-2000/model-00005-of-00005.safetensors +3 -0
- checkpoint-2000/model.safetensors.index.json +0 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/special_tokens_map.json +33 -0
- checkpoint-2000/tokenizer.json +3 -0
- checkpoint-2000/tokenizer.model +3 -0
- checkpoint-2000/tokenizer_config.json +0 -0
- checkpoint-2000/trainer_state.json +2474 -0
- checkpoint-2000/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-2000/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image_soft_token>": 262144
|
| 3 |
+
}
|
checkpoint-2000/chat_template.jinja
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{{ bos_token }}
|
| 2 |
+
{%- if messages[0]['role'] == 'system' -%}
|
| 3 |
+
{%- if messages[0]['content'] is string -%}
|
| 4 |
+
{%- set first_user_prefix = messages[0]['content'] + '
|
| 5 |
+
|
| 6 |
+
' -%}
|
| 7 |
+
{%- else -%}
|
| 8 |
+
{%- set first_user_prefix = messages[0]['content'][0]['text'] + '
|
| 9 |
+
|
| 10 |
+
' -%}
|
| 11 |
+
{%- endif -%}
|
| 12 |
+
{%- set loop_messages = messages[1:] -%}
|
| 13 |
+
{%- else -%}
|
| 14 |
+
{%- set first_user_prefix = "" -%}
|
| 15 |
+
{%- set loop_messages = messages -%}
|
| 16 |
+
{%- endif -%}
|
| 17 |
+
{%- for message in loop_messages -%}
|
| 18 |
+
{%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
|
| 19 |
+
{{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
|
| 20 |
+
{%- endif -%}
|
| 21 |
+
{%- if (message['role'] == 'assistant') -%}
|
| 22 |
+
{%- set role = "model" -%}
|
| 23 |
+
{%- else -%}
|
| 24 |
+
{%- set role = message['role'] -%}
|
| 25 |
+
{%- endif -%}
|
| 26 |
+
{{ '<start_of_turn>' + role + '
|
| 27 |
+
' + (first_user_prefix if loop.first else "") }}
|
| 28 |
+
{%- if message['content'] is string -%}
|
| 29 |
+
{{ message['content'] | trim }}
|
| 30 |
+
{%- elif message['content'] is iterable -%}
|
| 31 |
+
{%- for item in message['content'] -%}
|
| 32 |
+
{%- if item['type'] == 'image' -%}
|
| 33 |
+
{{ '<start_of_image>' }}
|
| 34 |
+
{%- elif item['type'] == 'text' -%}
|
| 35 |
+
{{ item['text'] | trim }}
|
| 36 |
+
{%- endif -%}
|
| 37 |
+
{%- endfor -%}
|
| 38 |
+
{%- else -%}
|
| 39 |
+
{{ raise_exception("Invalid content type") }}
|
| 40 |
+
{%- endif -%}
|
| 41 |
+
{{ '<end_of_turn>
|
| 42 |
+
' }}
|
| 43 |
+
{%- endfor -%}
|
| 44 |
+
{%- if add_generation_prompt -%}
|
| 45 |
+
{{'<start_of_turn>model
|
| 46 |
+
'}}
|
| 47 |
+
{%- endif -%}
|
checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma3ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"boi_token_index": 255999,
|
| 6 |
+
"bos_token_id": 2,
|
| 7 |
+
"dtype": "bfloat16",
|
| 8 |
+
"eoi_token_index": 256000,
|
| 9 |
+
"eos_token_id": 1,
|
| 10 |
+
"image_token_index": 262144,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"mm_tokens_per_image": 256,
|
| 13 |
+
"model_type": "gemma3",
|
| 14 |
+
"pad_token_id": 0,
|
| 15 |
+
"text_config": {
|
| 16 |
+
"_sliding_window_pattern": 6,
|
| 17 |
+
"attention_bias": false,
|
| 18 |
+
"attention_dropout": 0.0,
|
| 19 |
+
"attn_logit_softcapping": null,
|
| 20 |
+
"dtype": "bfloat16",
|
| 21 |
+
"final_logit_softcapping": null,
|
| 22 |
+
"head_dim": 256,
|
| 23 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 24 |
+
"hidden_size": 3840,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"intermediate_size": 15360,
|
| 27 |
+
"layer_types": [
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"sliding_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"sliding_attention",
|
| 38 |
+
"sliding_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"sliding_attention",
|
| 41 |
+
"sliding_attention",
|
| 42 |
+
"sliding_attention",
|
| 43 |
+
"sliding_attention",
|
| 44 |
+
"sliding_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"sliding_attention",
|
| 47 |
+
"sliding_attention",
|
| 48 |
+
"sliding_attention",
|
| 49 |
+
"sliding_attention",
|
| 50 |
+
"sliding_attention",
|
| 51 |
+
"full_attention",
|
| 52 |
+
"sliding_attention",
|
| 53 |
+
"sliding_attention",
|
| 54 |
+
"sliding_attention",
|
| 55 |
+
"sliding_attention",
|
| 56 |
+
"sliding_attention",
|
| 57 |
+
"full_attention",
|
| 58 |
+
"sliding_attention",
|
| 59 |
+
"sliding_attention",
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"sliding_attention",
|
| 63 |
+
"full_attention",
|
| 64 |
+
"sliding_attention",
|
| 65 |
+
"sliding_attention",
|
| 66 |
+
"sliding_attention",
|
| 67 |
+
"sliding_attention",
|
| 68 |
+
"sliding_attention",
|
| 69 |
+
"full_attention",
|
| 70 |
+
"sliding_attention",
|
| 71 |
+
"sliding_attention",
|
| 72 |
+
"sliding_attention",
|
| 73 |
+
"sliding_attention",
|
| 74 |
+
"sliding_attention",
|
| 75 |
+
"full_attention"
|
| 76 |
+
],
|
| 77 |
+
"max_position_embeddings": 131072,
|
| 78 |
+
"model_type": "gemma3_text",
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_hidden_layers": 48,
|
| 81 |
+
"num_key_value_heads": 8,
|
| 82 |
+
"query_pre_attn_scalar": 256,
|
| 83 |
+
"rms_norm_eps": 1e-06,
|
| 84 |
+
"rope_local_base_freq": 10000.0,
|
| 85 |
+
"rope_scaling": {
|
| 86 |
+
"factor": 8.0,
|
| 87 |
+
"rope_type": "linear"
|
| 88 |
+
},
|
| 89 |
+
"rope_theta": 1000000.0,
|
| 90 |
+
"sliding_window": 1024,
|
| 91 |
+
"use_cache": true,
|
| 92 |
+
"vocab_size": 262208
|
| 93 |
+
},
|
| 94 |
+
"transformers_version": "4.56.1",
|
| 95 |
+
"vision_config": {
|
| 96 |
+
"attention_dropout": 0.0,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"image_size": 896,
|
| 101 |
+
"intermediate_size": 4304,
|
| 102 |
+
"layer_norm_eps": 1e-06,
|
| 103 |
+
"model_type": "siglip_vision_model",
|
| 104 |
+
"num_attention_heads": 16,
|
| 105 |
+
"num_channels": 3,
|
| 106 |
+
"num_hidden_layers": 27,
|
| 107 |
+
"patch_size": 14,
|
| 108 |
+
"vision_use_head": false
|
| 109 |
+
}
|
| 110 |
+
}
|
checkpoint-2000/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 2,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
1,
|
| 6 |
+
1,
|
| 7 |
+
106
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 0,
|
| 10 |
+
"top_k": 64,
|
| 11 |
+
"top_p": 0.95,
|
| 12 |
+
"transformers_version": "4.56.1"
|
| 13 |
+
}
|
checkpoint-2000/model-00001-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4847447e92599833e8dbaa3067cd201c3bb5c052efa91f11ba891e43234f7832
|
| 3 |
+
size 4979902192
|
checkpoint-2000/model-00002-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:891bd54eed03cba9ee1e705533a02a8217fcc29f356e4a1f53e5fd0d178883ad
|
| 3 |
+
size 4931296592
|
checkpoint-2000/model-00003-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
|
| 3 |
+
size 4931296656
|
checkpoint-2000/model-00004-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c327d62bb136f97292c7f6cb7b6f356a7886f894c2307479c977f1300ec02dfc
|
| 3 |
+
size 4931296656
|
checkpoint-2000/model-00005-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:104f57108aaf19c277784c2d2cb0e9d4613b4bb1f8e85f357721b4791b93e2e6
|
| 3 |
+
size 4601000928
|
checkpoint-2000/model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a142c4e8891f8b510e486f9b2db27f483ed4ad3cdc196007b97713494a765f30
|
| 3 |
+
size 7286558019
|
checkpoint-2000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:039457aabfa4625f7b62a58eb9be9b588c4dd078fed6a4df33522308f5868222
|
| 3 |
+
size 14645
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:580ed3db3aadea223cc81ecfe2b5b0cc6b0a0c6e4530657b5b2acee6d9a72998
|
| 3 |
+
size 1465
|
checkpoint-2000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"boi_token": "<start_of_image>",
|
| 3 |
+
"bos_token": {
|
| 4 |
+
"content": "<bos>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
"eoi_token": "<end_of_image>",
|
| 11 |
+
"eos_token": {
|
| 12 |
+
"content": "<eos>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"image_token": "<image_soft_token>",
|
| 19 |
+
"pad_token": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false
|
| 25 |
+
},
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"content": "<unk>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": false,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
}
|
| 33 |
+
}
|
checkpoint-2000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
|
| 3 |
+
size 33384568
|
checkpoint-2000/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
|
| 3 |
+
size 4689074
|
checkpoint-2000/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,2474 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.2,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 2000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 0.7980934290215373,
|
| 14 |
+
"epoch": 0.016,
|
| 15 |
+
"grad_norm": 49.0,
|
| 16 |
+
"learning_rate": 1.8e-07,
|
| 17 |
+
"loss": 13.873,
|
| 18 |
+
"mean_token_accuracy": 0.77677470035851,
|
| 19 |
+
"num_tokens": 338352.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 0.7983348269015551,
|
| 24 |
+
"epoch": 0.032,
|
| 25 |
+
"grad_norm": 39.75,
|
| 26 |
+
"learning_rate": 3.8e-07,
|
| 27 |
+
"loss": 13.832,
|
| 28 |
+
"mean_token_accuracy": 0.7767552722245454,
|
| 29 |
+
"num_tokens": 671193.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 0.7940070297569036,
|
| 34 |
+
"epoch": 0.048,
|
| 35 |
+
"grad_norm": 36.5,
|
| 36 |
+
"learning_rate": 5.800000000000001e-07,
|
| 37 |
+
"loss": 13.6955,
|
| 38 |
+
"mean_token_accuracy": 0.7781575798988343,
|
| 39 |
+
"num_tokens": 1001008.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 0.8190817344933748,
|
| 44 |
+
"epoch": 0.064,
|
| 45 |
+
"grad_norm": 34.0,
|
| 46 |
+
"learning_rate": 7.8e-07,
|
| 47 |
+
"loss": 13.9863,
|
| 48 |
+
"mean_token_accuracy": 0.7736780602484942,
|
| 49 |
+
"num_tokens": 1331208.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 0.8228552235290408,
|
| 54 |
+
"epoch": 0.08,
|
| 55 |
+
"grad_norm": 33.0,
|
| 56 |
+
"learning_rate": 9.800000000000001e-07,
|
| 57 |
+
"loss": 13.7981,
|
| 58 |
+
"mean_token_accuracy": 0.7759026233106852,
|
| 59 |
+
"num_tokens": 1667532.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 0.8120608307421208,
|
| 64 |
+
"epoch": 0.096,
|
| 65 |
+
"grad_norm": 34.0,
|
| 66 |
+
"learning_rate": 1.1800000000000001e-06,
|
| 67 |
+
"loss": 13.4686,
|
| 68 |
+
"mean_token_accuracy": 0.7815129280090332,
|
| 69 |
+
"num_tokens": 2007176.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 0.8287483796477317,
|
| 74 |
+
"epoch": 0.112,
|
| 75 |
+
"grad_norm": 36.5,
|
| 76 |
+
"learning_rate": 1.3800000000000001e-06,
|
| 77 |
+
"loss": 13.7526,
|
| 78 |
+
"mean_token_accuracy": 0.776063147187233,
|
| 79 |
+
"num_tokens": 2342259.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 0.8556615479290486,
|
| 84 |
+
"epoch": 0.128,
|
| 85 |
+
"grad_norm": 27.125,
|
| 86 |
+
"learning_rate": 1.5800000000000001e-06,
|
| 87 |
+
"loss": 13.9267,
|
| 88 |
+
"mean_token_accuracy": 0.773414358869195,
|
| 89 |
+
"num_tokens": 2663987.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 0.8306544724851846,
|
| 94 |
+
"epoch": 0.144,
|
| 95 |
+
"grad_norm": 25.75,
|
| 96 |
+
"learning_rate": 1.7800000000000001e-06,
|
| 97 |
+
"loss": 13.6531,
|
| 98 |
+
"mean_token_accuracy": 0.7779791738837958,
|
| 99 |
+
"num_tokens": 2997824.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 0.8303449299186468,
|
| 104 |
+
"epoch": 0.16,
|
| 105 |
+
"grad_norm": 26.5,
|
| 106 |
+
"learning_rate": 1.98e-06,
|
| 107 |
+
"loss": 13.4855,
|
| 108 |
+
"mean_token_accuracy": 0.7814826928079128,
|
| 109 |
+
"num_tokens": 3330597.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.16,
|
| 114 |
+
"eval_biology_entropy": 1.125799958705902,
|
| 115 |
+
"eval_biology_loss": 1.1626057624816895,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.7041610617637635,
|
| 117 |
+
"eval_biology_num_tokens": 3330597.0,
|
| 118 |
+
"eval_biology_runtime": 29.5726,
|
| 119 |
+
"eval_biology_samples_per_second": 16.908,
|
| 120 |
+
"eval_biology_steps_per_second": 4.227,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.16,
|
| 125 |
+
"eval_physics_entropy": 0.8412163820266724,
|
| 126 |
+
"eval_physics_loss": 0.8517085909843445,
|
| 127 |
+
"eval_physics_mean_token_accuracy": 0.7784765930175781,
|
| 128 |
+
"eval_physics_num_tokens": 3330597.0,
|
| 129 |
+
"eval_physics_runtime": 42.3208,
|
| 130 |
+
"eval_physics_samples_per_second": 11.815,
|
| 131 |
+
"eval_physics_steps_per_second": 2.954,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 0.8291829921305179,
|
| 136 |
+
"epoch": 0.176,
|
| 137 |
+
"grad_norm": 32.5,
|
| 138 |
+
"learning_rate": 2.1800000000000003e-06,
|
| 139 |
+
"loss": 13.4483,
|
| 140 |
+
"mean_token_accuracy": 0.7799590565264225,
|
| 141 |
+
"num_tokens": 3658264.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 0.8301662255078555,
|
| 146 |
+
"epoch": 0.192,
|
| 147 |
+
"grad_norm": 25.5,
|
| 148 |
+
"learning_rate": 2.38e-06,
|
| 149 |
+
"loss": 13.4143,
|
| 150 |
+
"mean_token_accuracy": 0.7796540033072233,
|
| 151 |
+
"num_tokens": 3995568.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 0.821339176222682,
|
| 156 |
+
"epoch": 0.208,
|
| 157 |
+
"grad_norm": 26.125,
|
| 158 |
+
"learning_rate": 2.5800000000000003e-06,
|
| 159 |
+
"loss": 13.2097,
|
| 160 |
+
"mean_token_accuracy": 0.7838085852563381,
|
| 161 |
+
"num_tokens": 4321436.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 0.8153277274221182,
|
| 166 |
+
"epoch": 0.224,
|
| 167 |
+
"grad_norm": 24.0,
|
| 168 |
+
"learning_rate": 2.7800000000000005e-06,
|
| 169 |
+
"loss": 13.0671,
|
| 170 |
+
"mean_token_accuracy": 0.7830340649932623,
|
| 171 |
+
"num_tokens": 4648491.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 0.8157381068915128,
|
| 176 |
+
"epoch": 0.24,
|
| 177 |
+
"grad_norm": 23.375,
|
| 178 |
+
"learning_rate": 2.9800000000000003e-06,
|
| 179 |
+
"loss": 13.1148,
|
| 180 |
+
"mean_token_accuracy": 0.7832087766379118,
|
| 181 |
+
"num_tokens": 4986175.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"entropy": 0.8061813484877348,
|
| 186 |
+
"epoch": 0.256,
|
| 187 |
+
"grad_norm": 26.625,
|
| 188 |
+
"learning_rate": 3.1800000000000005e-06,
|
| 189 |
+
"loss": 13.0481,
|
| 190 |
+
"mean_token_accuracy": 0.7853319209069014,
|
| 191 |
+
"num_tokens": 5329320.0,
|
| 192 |
+
"step": 160
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"entropy": 0.8233318269252777,
|
| 196 |
+
"epoch": 0.272,
|
| 197 |
+
"grad_norm": 26.0,
|
| 198 |
+
"learning_rate": 3.3800000000000007e-06,
|
| 199 |
+
"loss": 13.1699,
|
| 200 |
+
"mean_token_accuracy": 0.7812638122588396,
|
| 201 |
+
"num_tokens": 5658796.0,
|
| 202 |
+
"step": 170
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"entropy": 0.8212542906403542,
|
| 206 |
+
"epoch": 0.288,
|
| 207 |
+
"grad_norm": 22.0,
|
| 208 |
+
"learning_rate": 3.58e-06,
|
| 209 |
+
"loss": 13.1426,
|
| 210 |
+
"mean_token_accuracy": 0.783899562805891,
|
| 211 |
+
"num_tokens": 5980201.0,
|
| 212 |
+
"step": 180
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"entropy": 0.8011876806616783,
|
| 216 |
+
"epoch": 0.304,
|
| 217 |
+
"grad_norm": 24.375,
|
| 218 |
+
"learning_rate": 3.7800000000000002e-06,
|
| 219 |
+
"loss": 12.8823,
|
| 220 |
+
"mean_token_accuracy": 0.7862063981592655,
|
| 221 |
+
"num_tokens": 6298858.0,
|
| 222 |
+
"step": 190
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"entropy": 0.8119899664074183,
|
| 226 |
+
"epoch": 0.32,
|
| 227 |
+
"grad_norm": 22.375,
|
| 228 |
+
"learning_rate": 3.980000000000001e-06,
|
| 229 |
+
"loss": 13.017,
|
| 230 |
+
"mean_token_accuracy": 0.7848218575119972,
|
| 231 |
+
"num_tokens": 6622798.0,
|
| 232 |
+
"step": 200
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.32,
|
| 236 |
+
"eval_biology_entropy": 1.137313835144043,
|
| 237 |
+
"eval_biology_loss": 1.1703258752822876,
|
| 238 |
+
"eval_biology_mean_token_accuracy": 0.7031810960769653,
|
| 239 |
+
"eval_biology_num_tokens": 6622798.0,
|
| 240 |
+
"eval_biology_runtime": 29.6247,
|
| 241 |
+
"eval_biology_samples_per_second": 16.878,
|
| 242 |
+
"eval_biology_steps_per_second": 4.219,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.32,
|
| 247 |
+
"eval_physics_entropy": 0.8116096572875977,
|
| 248 |
+
"eval_physics_loss": 0.8106478452682495,
|
| 249 |
+
"eval_physics_mean_token_accuracy": 0.7854588747024536,
|
| 250 |
+
"eval_physics_num_tokens": 6622798.0,
|
| 251 |
+
"eval_physics_runtime": 42.332,
|
| 252 |
+
"eval_physics_samples_per_second": 11.811,
|
| 253 |
+
"eval_physics_steps_per_second": 2.953,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"entropy": 0.8008256774395704,
|
| 258 |
+
"epoch": 0.336,
|
| 259 |
+
"grad_norm": 21.25,
|
| 260 |
+
"learning_rate": 4.18e-06,
|
| 261 |
+
"loss": 12.8337,
|
| 262 |
+
"mean_token_accuracy": 0.7872962314635515,
|
| 263 |
+
"num_tokens": 6953611.0,
|
| 264 |
+
"step": 210
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"entropy": 0.7870257034897804,
|
| 268 |
+
"epoch": 0.352,
|
| 269 |
+
"grad_norm": 23.75,
|
| 270 |
+
"learning_rate": 4.38e-06,
|
| 271 |
+
"loss": 12.629,
|
| 272 |
+
"mean_token_accuracy": 0.7910902902483941,
|
| 273 |
+
"num_tokens": 7290378.0,
|
| 274 |
+
"step": 220
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"entropy": 0.7753207640722394,
|
| 278 |
+
"epoch": 0.368,
|
| 279 |
+
"grad_norm": 23.625,
|
| 280 |
+
"learning_rate": 4.58e-06,
|
| 281 |
+
"loss": 12.3974,
|
| 282 |
+
"mean_token_accuracy": 0.7931954108178616,
|
| 283 |
+
"num_tokens": 7621082.0,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"entropy": 0.8035390684381127,
|
| 288 |
+
"epoch": 0.384,
|
| 289 |
+
"grad_norm": 23.375,
|
| 290 |
+
"learning_rate": 4.78e-06,
|
| 291 |
+
"loss": 12.8833,
|
| 292 |
+
"mean_token_accuracy": 0.7841386552900076,
|
| 293 |
+
"num_tokens": 7955570.0,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"entropy": 0.7771880365908146,
|
| 298 |
+
"epoch": 0.4,
|
| 299 |
+
"grad_norm": 23.5,
|
| 300 |
+
"learning_rate": 4.980000000000001e-06,
|
| 301 |
+
"loss": 12.462,
|
| 302 |
+
"mean_token_accuracy": 0.7913563769310713,
|
| 303 |
+
"num_tokens": 8291049.0,
|
| 304 |
+
"step": 250
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"entropy": 0.7649452563375234,
|
| 308 |
+
"epoch": 0.416,
|
| 309 |
+
"grad_norm": 22.25,
|
| 310 |
+
"learning_rate": 5.18e-06,
|
| 311 |
+
"loss": 12.2876,
|
| 312 |
+
"mean_token_accuracy": 0.7948728159070015,
|
| 313 |
+
"num_tokens": 8627310.0,
|
| 314 |
+
"step": 260
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"entropy": 0.7790930971503258,
|
| 318 |
+
"epoch": 0.432,
|
| 319 |
+
"grad_norm": 23.0,
|
| 320 |
+
"learning_rate": 5.380000000000001e-06,
|
| 321 |
+
"loss": 12.4896,
|
| 322 |
+
"mean_token_accuracy": 0.790573175624013,
|
| 323 |
+
"num_tokens": 8958371.0,
|
| 324 |
+
"step": 270
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"entropy": 0.7571947271004319,
|
| 328 |
+
"epoch": 0.448,
|
| 329 |
+
"grad_norm": 22.75,
|
| 330 |
+
"learning_rate": 5.580000000000001e-06,
|
| 331 |
+
"loss": 12.1797,
|
| 332 |
+
"mean_token_accuracy": 0.795942148193717,
|
| 333 |
+
"num_tokens": 9294388.0,
|
| 334 |
+
"step": 280
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"entropy": 0.7509586235508323,
|
| 338 |
+
"epoch": 0.464,
|
| 339 |
+
"grad_norm": 22.875,
|
| 340 |
+
"learning_rate": 5.78e-06,
|
| 341 |
+
"loss": 11.9835,
|
| 342 |
+
"mean_token_accuracy": 0.7969875477254391,
|
| 343 |
+
"num_tokens": 9620366.0,
|
| 344 |
+
"step": 290
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"entropy": 0.7548794999718667,
|
| 348 |
+
"epoch": 0.48,
|
| 349 |
+
"grad_norm": 20.875,
|
| 350 |
+
"learning_rate": 5.98e-06,
|
| 351 |
+
"loss": 12.0853,
|
| 352 |
+
"mean_token_accuracy": 0.7961488746106624,
|
| 353 |
+
"num_tokens": 9955431.0,
|
| 354 |
+
"step": 300
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.48,
|
| 358 |
+
"eval_biology_entropy": 1.1340459384918213,
|
| 359 |
+
"eval_biology_loss": 1.1766083240509033,
|
| 360 |
+
"eval_biology_mean_token_accuracy": 0.7020830793380737,
|
| 361 |
+
"eval_biology_num_tokens": 9955431.0,
|
| 362 |
+
"eval_biology_runtime": 29.8733,
|
| 363 |
+
"eval_biology_samples_per_second": 16.737,
|
| 364 |
+
"eval_biology_steps_per_second": 4.184,
|
| 365 |
+
"step": 300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.48,
|
| 369 |
+
"eval_physics_entropy": 0.7825248126983643,
|
| 370 |
+
"eval_physics_loss": 0.7770141363143921,
|
| 371 |
+
"eval_physics_mean_token_accuracy": 0.7918553309440612,
|
| 372 |
+
"eval_physics_num_tokens": 9955431.0,
|
| 373 |
+
"eval_physics_runtime": 42.3494,
|
| 374 |
+
"eval_physics_samples_per_second": 11.807,
|
| 375 |
+
"eval_physics_steps_per_second": 2.952,
|
| 376 |
+
"step": 300
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"entropy": 0.7715075328946114,
|
| 380 |
+
"epoch": 0.496,
|
| 381 |
+
"grad_norm": 20.375,
|
| 382 |
+
"learning_rate": 6.18e-06,
|
| 383 |
+
"loss": 12.3217,
|
| 384 |
+
"mean_token_accuracy": 0.7934962294995784,
|
| 385 |
+
"num_tokens": 10289883.0,
|
| 386 |
+
"step": 310
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"entropy": 0.7606870967894792,
|
| 390 |
+
"epoch": 0.512,
|
| 391 |
+
"grad_norm": 24.625,
|
| 392 |
+
"learning_rate": 6.380000000000001e-06,
|
| 393 |
+
"loss": 12.2158,
|
| 394 |
+
"mean_token_accuracy": 0.795098090544343,
|
| 395 |
+
"num_tokens": 10619468.0,
|
| 396 |
+
"step": 320
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"entropy": 0.7731524888426066,
|
| 400 |
+
"epoch": 0.528,
|
| 401 |
+
"grad_norm": 20.5,
|
| 402 |
+
"learning_rate": 6.5800000000000005e-06,
|
| 403 |
+
"loss": 12.3882,
|
| 404 |
+
"mean_token_accuracy": 0.7916440792381764,
|
| 405 |
+
"num_tokens": 10944904.0,
|
| 406 |
+
"step": 330
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"entropy": 0.755775578506291,
|
| 410 |
+
"epoch": 0.544,
|
| 411 |
+
"grad_norm": 22.875,
|
| 412 |
+
"learning_rate": 6.780000000000001e-06,
|
| 413 |
+
"loss": 12.1334,
|
| 414 |
+
"mean_token_accuracy": 0.7953360237181186,
|
| 415 |
+
"num_tokens": 11266409.0,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"entropy": 0.7537049597129226,
|
| 420 |
+
"epoch": 0.56,
|
| 421 |
+
"grad_norm": 18.875,
|
| 422 |
+
"learning_rate": 6.98e-06,
|
| 423 |
+
"loss": 12.0364,
|
| 424 |
+
"mean_token_accuracy": 0.7975195907056332,
|
| 425 |
+
"num_tokens": 11605544.0,
|
| 426 |
+
"step": 350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"entropy": 0.7410756940022111,
|
| 430 |
+
"epoch": 0.576,
|
| 431 |
+
"grad_norm": 21.375,
|
| 432 |
+
"learning_rate": 7.180000000000001e-06,
|
| 433 |
+
"loss": 11.8459,
|
| 434 |
+
"mean_token_accuracy": 0.8007426861673593,
|
| 435 |
+
"num_tokens": 11931715.0,
|
| 436 |
+
"step": 360
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"entropy": 0.7542769065126776,
|
| 440 |
+
"epoch": 0.592,
|
| 441 |
+
"grad_norm": 21.375,
|
| 442 |
+
"learning_rate": 7.3800000000000005e-06,
|
| 443 |
+
"loss": 12.0135,
|
| 444 |
+
"mean_token_accuracy": 0.7966224331408739,
|
| 445 |
+
"num_tokens": 12257243.0,
|
| 446 |
+
"step": 370
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"entropy": 0.7349719753488898,
|
| 450 |
+
"epoch": 0.608,
|
| 451 |
+
"grad_norm": 20.375,
|
| 452 |
+
"learning_rate": 7.58e-06,
|
| 453 |
+
"loss": 11.7827,
|
| 454 |
+
"mean_token_accuracy": 0.7996757850050926,
|
| 455 |
+
"num_tokens": 12583154.0,
|
| 456 |
+
"step": 380
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"entropy": 0.7510534537956118,
|
| 460 |
+
"epoch": 0.624,
|
| 461 |
+
"grad_norm": 20.875,
|
| 462 |
+
"learning_rate": 7.78e-06,
|
| 463 |
+
"loss": 12.0433,
|
| 464 |
+
"mean_token_accuracy": 0.7961709950119257,
|
| 465 |
+
"num_tokens": 12905392.0,
|
| 466 |
+
"step": 390
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"entropy": 0.746537908166647,
|
| 470 |
+
"epoch": 0.64,
|
| 471 |
+
"grad_norm": 18.75,
|
| 472 |
+
"learning_rate": 7.980000000000002e-06,
|
| 473 |
+
"loss": 11.9585,
|
| 474 |
+
"mean_token_accuracy": 0.798334264755249,
|
| 475 |
+
"num_tokens": 13232198.0,
|
| 476 |
+
"step": 400
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.64,
|
| 480 |
+
"eval_biology_entropy": 1.100157816886902,
|
| 481 |
+
"eval_biology_loss": 1.1817364692687988,
|
| 482 |
+
"eval_biology_mean_token_accuracy": 0.7015948586463928,
|
| 483 |
+
"eval_biology_num_tokens": 13232198.0,
|
| 484 |
+
"eval_biology_runtime": 29.5996,
|
| 485 |
+
"eval_biology_samples_per_second": 16.892,
|
| 486 |
+
"eval_biology_steps_per_second": 4.223,
|
| 487 |
+
"step": 400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.64,
|
| 491 |
+
"eval_physics_entropy": 0.7428050627708435,
|
| 492 |
+
"eval_physics_loss": 0.7491776943206787,
|
| 493 |
+
"eval_physics_mean_token_accuracy": 0.7974028449058532,
|
| 494 |
+
"eval_physics_num_tokens": 13232198.0,
|
| 495 |
+
"eval_physics_runtime": 42.3622,
|
| 496 |
+
"eval_physics_samples_per_second": 11.803,
|
| 497 |
+
"eval_physics_steps_per_second": 2.951,
|
| 498 |
+
"step": 400
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"entropy": 0.7335962850600481,
|
| 502 |
+
"epoch": 0.656,
|
| 503 |
+
"grad_norm": 18.125,
|
| 504 |
+
"learning_rate": 8.18e-06,
|
| 505 |
+
"loss": 11.7731,
|
| 506 |
+
"mean_token_accuracy": 0.8001615725457668,
|
| 507 |
+
"num_tokens": 13575902.0,
|
| 508 |
+
"step": 410
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"entropy": 0.7217804849147796,
|
| 512 |
+
"epoch": 0.672,
|
| 513 |
+
"grad_norm": 21.25,
|
| 514 |
+
"learning_rate": 8.380000000000001e-06,
|
| 515 |
+
"loss": 11.5198,
|
| 516 |
+
"mean_token_accuracy": 0.8043914053589105,
|
| 517 |
+
"num_tokens": 13895997.0,
|
| 518 |
+
"step": 420
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"entropy": 0.7267325738444924,
|
| 522 |
+
"epoch": 0.688,
|
| 523 |
+
"grad_norm": 20.75,
|
| 524 |
+
"learning_rate": 8.580000000000001e-06,
|
| 525 |
+
"loss": 11.6352,
|
| 526 |
+
"mean_token_accuracy": 0.8020107671618462,
|
| 527 |
+
"num_tokens": 14234888.0,
|
| 528 |
+
"step": 430
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"entropy": 0.741226227581501,
|
| 532 |
+
"epoch": 0.704,
|
| 533 |
+
"grad_norm": 21.5,
|
| 534 |
+
"learning_rate": 8.78e-06,
|
| 535 |
+
"loss": 11.9022,
|
| 536 |
+
"mean_token_accuracy": 0.7984758585691452,
|
| 537 |
+
"num_tokens": 14567908.0,
|
| 538 |
+
"step": 440
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"entropy": 0.722562575340271,
|
| 542 |
+
"epoch": 0.72,
|
| 543 |
+
"grad_norm": 22.875,
|
| 544 |
+
"learning_rate": 8.98e-06,
|
| 545 |
+
"loss": 11.5666,
|
| 546 |
+
"mean_token_accuracy": 0.8033338606357574,
|
| 547 |
+
"num_tokens": 14882927.0,
|
| 548 |
+
"step": 450
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"entropy": 0.7157323375344277,
|
| 552 |
+
"epoch": 0.736,
|
| 553 |
+
"grad_norm": 22.25,
|
| 554 |
+
"learning_rate": 9.180000000000002e-06,
|
| 555 |
+
"loss": 11.464,
|
| 556 |
+
"mean_token_accuracy": 0.8032864931970835,
|
| 557 |
+
"num_tokens": 15217342.0,
|
| 558 |
+
"step": 460
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"entropy": 0.710782004147768,
|
| 562 |
+
"epoch": 0.752,
|
| 563 |
+
"grad_norm": 18.5,
|
| 564 |
+
"learning_rate": 9.38e-06,
|
| 565 |
+
"loss": 11.3824,
|
| 566 |
+
"mean_token_accuracy": 0.8050779778510332,
|
| 567 |
+
"num_tokens": 15547710.0,
|
| 568 |
+
"step": 470
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"entropy": 0.7326425770297647,
|
| 572 |
+
"epoch": 0.768,
|
| 573 |
+
"grad_norm": 20.375,
|
| 574 |
+
"learning_rate": 9.58e-06,
|
| 575 |
+
"loss": 11.7364,
|
| 576 |
+
"mean_token_accuracy": 0.7999915853142738,
|
| 577 |
+
"num_tokens": 15877177.0,
|
| 578 |
+
"step": 480
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"entropy": 0.7081343689933419,
|
| 582 |
+
"epoch": 0.784,
|
| 583 |
+
"grad_norm": 18.375,
|
| 584 |
+
"learning_rate": 9.780000000000001e-06,
|
| 585 |
+
"loss": 11.3161,
|
| 586 |
+
"mean_token_accuracy": 0.8068799041211605,
|
| 587 |
+
"num_tokens": 16219640.0,
|
| 588 |
+
"step": 490
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"entropy": 0.7145562319085002,
|
| 592 |
+
"epoch": 0.8,
|
| 593 |
+
"grad_norm": 19.375,
|
| 594 |
+
"learning_rate": 9.980000000000001e-06,
|
| 595 |
+
"loss": 11.4904,
|
| 596 |
+
"mean_token_accuracy": 0.8040819689631462,
|
| 597 |
+
"num_tokens": 16548261.0,
|
| 598 |
+
"step": 500
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.8,
|
| 602 |
+
"eval_biology_entropy": 1.1381474986076354,
|
| 603 |
+
"eval_biology_loss": 1.1847245693206787,
|
| 604 |
+
"eval_biology_mean_token_accuracy": 0.6994160051345826,
|
| 605 |
+
"eval_biology_num_tokens": 16548261.0,
|
| 606 |
+
"eval_biology_runtime": 29.6161,
|
| 607 |
+
"eval_biology_samples_per_second": 16.883,
|
| 608 |
+
"eval_biology_steps_per_second": 4.221,
|
| 609 |
+
"step": 500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.8,
|
| 613 |
+
"eval_physics_entropy": 0.7353026342391967,
|
| 614 |
+
"eval_physics_loss": 0.7241687178611755,
|
| 615 |
+
"eval_physics_mean_token_accuracy": 0.8026187562942505,
|
| 616 |
+
"eval_physics_num_tokens": 16548261.0,
|
| 617 |
+
"eval_physics_runtime": 42.1602,
|
| 618 |
+
"eval_physics_samples_per_second": 11.86,
|
| 619 |
+
"eval_physics_steps_per_second": 2.965,
|
| 620 |
+
"step": 500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"entropy": 0.715489343367517,
|
| 624 |
+
"epoch": 0.816,
|
| 625 |
+
"grad_norm": 18.75,
|
| 626 |
+
"learning_rate": 1.018e-05,
|
| 627 |
+
"loss": 11.4335,
|
| 628 |
+
"mean_token_accuracy": 0.8032866880297661,
|
| 629 |
+
"num_tokens": 16870533.0,
|
| 630 |
+
"step": 510
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"entropy": 0.7197281535714865,
|
| 634 |
+
"epoch": 0.832,
|
| 635 |
+
"grad_norm": 22.625,
|
| 636 |
+
"learning_rate": 1.038e-05,
|
| 637 |
+
"loss": 11.5253,
|
| 638 |
+
"mean_token_accuracy": 0.803961057588458,
|
| 639 |
+
"num_tokens": 17196371.0,
|
| 640 |
+
"step": 520
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"entropy": 0.7022356629371643,
|
| 644 |
+
"epoch": 0.848,
|
| 645 |
+
"grad_norm": 17.375,
|
| 646 |
+
"learning_rate": 1.0580000000000002e-05,
|
| 647 |
+
"loss": 11.2777,
|
| 648 |
+
"mean_token_accuracy": 0.8062990486621857,
|
| 649 |
+
"num_tokens": 17527066.0,
|
| 650 |
+
"step": 530
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"entropy": 0.7121493600308895,
|
| 654 |
+
"epoch": 0.864,
|
| 655 |
+
"grad_norm": 17.625,
|
| 656 |
+
"learning_rate": 1.0780000000000002e-05,
|
| 657 |
+
"loss": 11.3616,
|
| 658 |
+
"mean_token_accuracy": 0.8058860942721366,
|
| 659 |
+
"num_tokens": 17859605.0,
|
| 660 |
+
"step": 540
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"entropy": 0.7143916061148048,
|
| 664 |
+
"epoch": 0.88,
|
| 665 |
+
"grad_norm": 19.875,
|
| 666 |
+
"learning_rate": 1.0980000000000002e-05,
|
| 667 |
+
"loss": 11.4128,
|
| 668 |
+
"mean_token_accuracy": 0.8041808195412159,
|
| 669 |
+
"num_tokens": 18187315.0,
|
| 670 |
+
"step": 550
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"entropy": 0.6755495458841324,
|
| 674 |
+
"epoch": 0.896,
|
| 675 |
+
"grad_norm": 17.375,
|
| 676 |
+
"learning_rate": 1.1180000000000001e-05,
|
| 677 |
+
"loss": 10.8244,
|
| 678 |
+
"mean_token_accuracy": 0.8135501727461815,
|
| 679 |
+
"num_tokens": 18536790.0,
|
| 680 |
+
"step": 560
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"entropy": 0.7111412728205323,
|
| 684 |
+
"epoch": 0.912,
|
| 685 |
+
"grad_norm": 18.5,
|
| 686 |
+
"learning_rate": 1.138e-05,
|
| 687 |
+
"loss": 11.3738,
|
| 688 |
+
"mean_token_accuracy": 0.8046688307076693,
|
| 689 |
+
"num_tokens": 18869344.0,
|
| 690 |
+
"step": 570
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"entropy": 0.7016011888161302,
|
| 694 |
+
"epoch": 0.928,
|
| 695 |
+
"grad_norm": 20.75,
|
| 696 |
+
"learning_rate": 1.1580000000000001e-05,
|
| 697 |
+
"loss": 11.2569,
|
| 698 |
+
"mean_token_accuracy": 0.8068617224693299,
|
| 699 |
+
"num_tokens": 19199834.0,
|
| 700 |
+
"step": 580
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"entropy": 0.6901180900633335,
|
| 704 |
+
"epoch": 0.944,
|
| 705 |
+
"grad_norm": 19.375,
|
| 706 |
+
"learning_rate": 1.178e-05,
|
| 707 |
+
"loss": 11.0197,
|
| 708 |
+
"mean_token_accuracy": 0.8105497155338526,
|
| 709 |
+
"num_tokens": 19525385.0,
|
| 710 |
+
"step": 590
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"entropy": 0.6980544172227383,
|
| 714 |
+
"epoch": 0.96,
|
| 715 |
+
"grad_norm": 18.0,
|
| 716 |
+
"learning_rate": 1.198e-05,
|
| 717 |
+
"loss": 11.1812,
|
| 718 |
+
"mean_token_accuracy": 0.8075801994651556,
|
| 719 |
+
"num_tokens": 19856973.0,
|
| 720 |
+
"step": 600
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.96,
|
| 724 |
+
"eval_biology_entropy": 1.10544912815094,
|
| 725 |
+
"eval_biology_loss": 1.1910302639007568,
|
| 726 |
+
"eval_biology_mean_token_accuracy": 0.7007143182754516,
|
| 727 |
+
"eval_biology_num_tokens": 19856973.0,
|
| 728 |
+
"eval_biology_runtime": 29.6701,
|
| 729 |
+
"eval_biology_samples_per_second": 16.852,
|
| 730 |
+
"eval_biology_steps_per_second": 4.213,
|
| 731 |
+
"step": 600
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.96,
|
| 735 |
+
"eval_physics_entropy": 0.7146178777217865,
|
| 736 |
+
"eval_physics_loss": 0.7040261030197144,
|
| 737 |
+
"eval_physics_mean_token_accuracy": 0.8063621273040772,
|
| 738 |
+
"eval_physics_num_tokens": 19856973.0,
|
| 739 |
+
"eval_physics_runtime": 42.3789,
|
| 740 |
+
"eval_physics_samples_per_second": 11.798,
|
| 741 |
+
"eval_physics_steps_per_second": 2.95,
|
| 742 |
+
"step": 600
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"entropy": 0.6889748936519027,
|
| 746 |
+
"epoch": 0.976,
|
| 747 |
+
"grad_norm": 18.375,
|
| 748 |
+
"learning_rate": 1.218e-05,
|
| 749 |
+
"loss": 10.9979,
|
| 750 |
+
"mean_token_accuracy": 0.8105814348906278,
|
| 751 |
+
"num_tokens": 20190600.0,
|
| 752 |
+
"step": 610
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"entropy": 0.6845765860751272,
|
| 756 |
+
"epoch": 0.992,
|
| 757 |
+
"grad_norm": 17.875,
|
| 758 |
+
"learning_rate": 1.2380000000000002e-05,
|
| 759 |
+
"loss": 10.9961,
|
| 760 |
+
"mean_token_accuracy": 0.8099663183093071,
|
| 761 |
+
"num_tokens": 20518014.0,
|
| 762 |
+
"step": 620
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"entropy": 0.6710353532806039,
|
| 766 |
+
"epoch": 1.008,
|
| 767 |
+
"grad_norm": 17.375,
|
| 768 |
+
"learning_rate": 1.2580000000000002e-05,
|
| 769 |
+
"loss": 10.6734,
|
| 770 |
+
"mean_token_accuracy": 0.8143552124500275,
|
| 771 |
+
"num_tokens": 20859215.0,
|
| 772 |
+
"step": 630
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"entropy": 0.671150723285973,
|
| 776 |
+
"epoch": 1.024,
|
| 777 |
+
"grad_norm": 18.875,
|
| 778 |
+
"learning_rate": 1.2780000000000001e-05,
|
| 779 |
+
"loss": 10.7577,
|
| 780 |
+
"mean_token_accuracy": 0.8129799973219634,
|
| 781 |
+
"num_tokens": 21186658.0,
|
| 782 |
+
"step": 640
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"entropy": 0.670030782930553,
|
| 786 |
+
"epoch": 1.04,
|
| 787 |
+
"grad_norm": 17.75,
|
| 788 |
+
"learning_rate": 1.2980000000000001e-05,
|
| 789 |
+
"loss": 10.7736,
|
| 790 |
+
"mean_token_accuracy": 0.8133355241268874,
|
| 791 |
+
"num_tokens": 21517237.0,
|
| 792 |
+
"step": 650
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"entropy": 0.6583073100075125,
|
| 796 |
+
"epoch": 1.056,
|
| 797 |
+
"grad_norm": 16.75,
|
| 798 |
+
"learning_rate": 1.3180000000000001e-05,
|
| 799 |
+
"loss": 10.521,
|
| 800 |
+
"mean_token_accuracy": 0.8164310045540333,
|
| 801 |
+
"num_tokens": 21852528.0,
|
| 802 |
+
"step": 660
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"entropy": 0.67260313462466,
|
| 806 |
+
"epoch": 1.072,
|
| 807 |
+
"grad_norm": 17.875,
|
| 808 |
+
"learning_rate": 1.3380000000000002e-05,
|
| 809 |
+
"loss": 10.711,
|
| 810 |
+
"mean_token_accuracy": 0.8140104580670595,
|
| 811 |
+
"num_tokens": 22192380.0,
|
| 812 |
+
"step": 670
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"entropy": 0.6687374956905842,
|
| 816 |
+
"epoch": 1.088,
|
| 817 |
+
"grad_norm": 20.625,
|
| 818 |
+
"learning_rate": 1.3580000000000002e-05,
|
| 819 |
+
"loss": 10.7304,
|
| 820 |
+
"mean_token_accuracy": 0.8136457096785307,
|
| 821 |
+
"num_tokens": 22517971.0,
|
| 822 |
+
"step": 680
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"entropy": 0.6568242866545916,
|
| 826 |
+
"epoch": 1.104,
|
| 827 |
+
"grad_norm": 16.5,
|
| 828 |
+
"learning_rate": 1.378e-05,
|
| 829 |
+
"loss": 10.5146,
|
| 830 |
+
"mean_token_accuracy": 0.8156068984419107,
|
| 831 |
+
"num_tokens": 22848694.0,
|
| 832 |
+
"step": 690
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"entropy": 0.6624812703579664,
|
| 836 |
+
"epoch": 1.12,
|
| 837 |
+
"grad_norm": 18.25,
|
| 838 |
+
"learning_rate": 1.398e-05,
|
| 839 |
+
"loss": 10.6908,
|
| 840 |
+
"mean_token_accuracy": 0.8136210646480322,
|
| 841 |
+
"num_tokens": 23183889.0,
|
| 842 |
+
"step": 700
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.12,
|
| 846 |
+
"eval_biology_entropy": 1.096513487815857,
|
| 847 |
+
"eval_biology_loss": 1.1950488090515137,
|
| 848 |
+
"eval_biology_mean_token_accuracy": 0.6985604496002197,
|
| 849 |
+
"eval_biology_num_tokens": 23183889.0,
|
| 850 |
+
"eval_biology_runtime": 29.5973,
|
| 851 |
+
"eval_biology_samples_per_second": 16.893,
|
| 852 |
+
"eval_biology_steps_per_second": 4.223,
|
| 853 |
+
"step": 700
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 1.12,
|
| 857 |
+
"eval_physics_entropy": 0.6853570246696472,
|
| 858 |
+
"eval_physics_loss": 0.6883610486984253,
|
| 859 |
+
"eval_physics_mean_token_accuracy": 0.8095321311950684,
|
| 860 |
+
"eval_physics_num_tokens": 23183889.0,
|
| 861 |
+
"eval_physics_runtime": 42.3537,
|
| 862 |
+
"eval_physics_samples_per_second": 11.805,
|
| 863 |
+
"eval_physics_steps_per_second": 2.951,
|
| 864 |
+
"step": 700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"entropy": 0.6393926108255983,
|
| 868 |
+
"epoch": 1.1360000000000001,
|
| 869 |
+
"grad_norm": 17.125,
|
| 870 |
+
"learning_rate": 1.418e-05,
|
| 871 |
+
"loss": 10.2691,
|
| 872 |
+
"mean_token_accuracy": 0.8204710930585861,
|
| 873 |
+
"num_tokens": 23523903.0,
|
| 874 |
+
"step": 710
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"entropy": 0.6592076497152448,
|
| 878 |
+
"epoch": 1.152,
|
| 879 |
+
"grad_norm": 19.5,
|
| 880 |
+
"learning_rate": 1.4380000000000001e-05,
|
| 881 |
+
"loss": 10.525,
|
| 882 |
+
"mean_token_accuracy": 0.8170250646770001,
|
| 883 |
+
"num_tokens": 23848848.0,
|
| 884 |
+
"step": 720
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"entropy": 0.6672398639842868,
|
| 888 |
+
"epoch": 1.168,
|
| 889 |
+
"grad_norm": 19.25,
|
| 890 |
+
"learning_rate": 1.4580000000000001e-05,
|
| 891 |
+
"loss": 10.7446,
|
| 892 |
+
"mean_token_accuracy": 0.8126429941505193,
|
| 893 |
+
"num_tokens": 24174589.0,
|
| 894 |
+
"step": 730
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"entropy": 0.6579326229169965,
|
| 898 |
+
"epoch": 1.184,
|
| 899 |
+
"grad_norm": 20.0,
|
| 900 |
+
"learning_rate": 1.478e-05,
|
| 901 |
+
"loss": 10.5484,
|
| 902 |
+
"mean_token_accuracy": 0.8169561930000782,
|
| 903 |
+
"num_tokens": 24495981.0,
|
| 904 |
+
"step": 740
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"entropy": 0.6486070465296507,
|
| 908 |
+
"epoch": 1.2,
|
| 909 |
+
"grad_norm": 17.875,
|
| 910 |
+
"learning_rate": 1.498e-05,
|
| 911 |
+
"loss": 10.4176,
|
| 912 |
+
"mean_token_accuracy": 0.818020261451602,
|
| 913 |
+
"num_tokens": 24826245.0,
|
| 914 |
+
"step": 750
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"entropy": 0.6617947239428759,
|
| 918 |
+
"epoch": 1.216,
|
| 919 |
+
"grad_norm": 19.25,
|
| 920 |
+
"learning_rate": 1.5180000000000002e-05,
|
| 921 |
+
"loss": 10.5833,
|
| 922 |
+
"mean_token_accuracy": 0.8140788938850164,
|
| 923 |
+
"num_tokens": 25147775.0,
|
| 924 |
+
"step": 760
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"entropy": 0.6527341572567821,
|
| 928 |
+
"epoch": 1.232,
|
| 929 |
+
"grad_norm": 17.875,
|
| 930 |
+
"learning_rate": 1.5380000000000002e-05,
|
| 931 |
+
"loss": 10.4663,
|
| 932 |
+
"mean_token_accuracy": 0.8178805992007255,
|
| 933 |
+
"num_tokens": 25480033.0,
|
| 934 |
+
"step": 770
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"entropy": 0.6440409064292908,
|
| 938 |
+
"epoch": 1.248,
|
| 939 |
+
"grad_norm": 18.25,
|
| 940 |
+
"learning_rate": 1.5580000000000003e-05,
|
| 941 |
+
"loss": 10.3188,
|
| 942 |
+
"mean_token_accuracy": 0.819524721801281,
|
| 943 |
+
"num_tokens": 25811978.0,
|
| 944 |
+
"step": 780
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"entropy": 0.6459297401830554,
|
| 948 |
+
"epoch": 1.264,
|
| 949 |
+
"grad_norm": 17.5,
|
| 950 |
+
"learning_rate": 1.578e-05,
|
| 951 |
+
"loss": 10.3328,
|
| 952 |
+
"mean_token_accuracy": 0.8194583039730787,
|
| 953 |
+
"num_tokens": 26146299.0,
|
| 954 |
+
"step": 790
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"entropy": 0.6661811180412769,
|
| 958 |
+
"epoch": 1.28,
|
| 959 |
+
"grad_norm": 18.5,
|
| 960 |
+
"learning_rate": 1.5980000000000003e-05,
|
| 961 |
+
"loss": 10.6612,
|
| 962 |
+
"mean_token_accuracy": 0.813135676458478,
|
| 963 |
+
"num_tokens": 26478207.0,
|
| 964 |
+
"step": 800
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.28,
|
| 968 |
+
"eval_biology_entropy": 1.0779736218452454,
|
| 969 |
+
"eval_biology_loss": 1.1969928741455078,
|
| 970 |
+
"eval_biology_mean_token_accuracy": 0.6996237397193908,
|
| 971 |
+
"eval_biology_num_tokens": 26478207.0,
|
| 972 |
+
"eval_biology_runtime": 29.6158,
|
| 973 |
+
"eval_biology_samples_per_second": 16.883,
|
| 974 |
+
"eval_biology_steps_per_second": 4.221,
|
| 975 |
+
"step": 800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.28,
|
| 979 |
+
"eval_physics_entropy": 0.6634284896850586,
|
| 980 |
+
"eval_physics_loss": 0.6750168800354004,
|
| 981 |
+
"eval_physics_mean_token_accuracy": 0.8123881487846375,
|
| 982 |
+
"eval_physics_num_tokens": 26478207.0,
|
| 983 |
+
"eval_physics_runtime": 42.3789,
|
| 984 |
+
"eval_physics_samples_per_second": 11.798,
|
| 985 |
+
"eval_physics_steps_per_second": 2.95,
|
| 986 |
+
"step": 800
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"entropy": 0.6371305795386434,
|
| 990 |
+
"epoch": 1.296,
|
| 991 |
+
"grad_norm": 18.375,
|
| 992 |
+
"learning_rate": 1.618e-05,
|
| 993 |
+
"loss": 10.2353,
|
| 994 |
+
"mean_token_accuracy": 0.8211836166679859,
|
| 995 |
+
"num_tokens": 26806477.0,
|
| 996 |
+
"step": 810
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"entropy": 0.6579740369692445,
|
| 1000 |
+
"epoch": 1.312,
|
| 1001 |
+
"grad_norm": 18.5,
|
| 1002 |
+
"learning_rate": 1.638e-05,
|
| 1003 |
+
"loss": 10.5754,
|
| 1004 |
+
"mean_token_accuracy": 0.8152522802352905,
|
| 1005 |
+
"num_tokens": 27137528.0,
|
| 1006 |
+
"step": 820
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"entropy": 0.6427882853895426,
|
| 1010 |
+
"epoch": 1.328,
|
| 1011 |
+
"grad_norm": 18.5,
|
| 1012 |
+
"learning_rate": 1.658e-05,
|
| 1013 |
+
"loss": 10.3247,
|
| 1014 |
+
"mean_token_accuracy": 0.8178475011140108,
|
| 1015 |
+
"num_tokens": 27465859.0,
|
| 1016 |
+
"step": 830
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"entropy": 0.6479509010910988,
|
| 1020 |
+
"epoch": 1.3439999999999999,
|
| 1021 |
+
"grad_norm": 19.0,
|
| 1022 |
+
"learning_rate": 1.6780000000000002e-05,
|
| 1023 |
+
"loss": 10.4341,
|
| 1024 |
+
"mean_token_accuracy": 0.8170580971986056,
|
| 1025 |
+
"num_tokens": 27795568.0,
|
| 1026 |
+
"step": 840
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"entropy": 0.6500799937173725,
|
| 1030 |
+
"epoch": 1.3599999999999999,
|
| 1031 |
+
"grad_norm": 15.9375,
|
| 1032 |
+
"learning_rate": 1.698e-05,
|
| 1033 |
+
"loss": 10.383,
|
| 1034 |
+
"mean_token_accuracy": 0.8165912210941315,
|
| 1035 |
+
"num_tokens": 28129988.0,
|
| 1036 |
+
"step": 850
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"entropy": 0.6481674930080772,
|
| 1040 |
+
"epoch": 1.376,
|
| 1041 |
+
"grad_norm": 20.125,
|
| 1042 |
+
"learning_rate": 1.718e-05,
|
| 1043 |
+
"loss": 10.3763,
|
| 1044 |
+
"mean_token_accuracy": 0.8186531469225884,
|
| 1045 |
+
"num_tokens": 28465903.0,
|
| 1046 |
+
"step": 860
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"entropy": 0.6292730543762446,
|
| 1050 |
+
"epoch": 1.392,
|
| 1051 |
+
"grad_norm": 17.0,
|
| 1052 |
+
"learning_rate": 1.7380000000000003e-05,
|
| 1053 |
+
"loss": 10.0886,
|
| 1054 |
+
"mean_token_accuracy": 0.8228901036083698,
|
| 1055 |
+
"num_tokens": 28801554.0,
|
| 1056 |
+
"step": 870
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"entropy": 0.6366163183003664,
|
| 1060 |
+
"epoch": 1.408,
|
| 1061 |
+
"grad_norm": 18.375,
|
| 1062 |
+
"learning_rate": 1.758e-05,
|
| 1063 |
+
"loss": 10.2299,
|
| 1064 |
+
"mean_token_accuracy": 0.8191725321114063,
|
| 1065 |
+
"num_tokens": 29125114.0,
|
| 1066 |
+
"step": 880
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"entropy": 0.6577673699706793,
|
| 1070 |
+
"epoch": 1.424,
|
| 1071 |
+
"grad_norm": 18.75,
|
| 1072 |
+
"learning_rate": 1.7780000000000003e-05,
|
| 1073 |
+
"loss": 10.5993,
|
| 1074 |
+
"mean_token_accuracy": 0.8148027937859297,
|
| 1075 |
+
"num_tokens": 29454334.0,
|
| 1076 |
+
"step": 890
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"entropy": 0.629906609095633,
|
| 1080 |
+
"epoch": 1.44,
|
| 1081 |
+
"grad_norm": 17.25,
|
| 1082 |
+
"learning_rate": 1.798e-05,
|
| 1083 |
+
"loss": 10.0722,
|
| 1084 |
+
"mean_token_accuracy": 0.8225096724927425,
|
| 1085 |
+
"num_tokens": 29789253.0,
|
| 1086 |
+
"step": 900
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 1.44,
|
| 1090 |
+
"eval_biology_entropy": 1.0743142724037171,
|
| 1091 |
+
"eval_biology_loss": 1.2018756866455078,
|
| 1092 |
+
"eval_biology_mean_token_accuracy": 0.6978029899597168,
|
| 1093 |
+
"eval_biology_num_tokens": 29789253.0,
|
| 1094 |
+
"eval_biology_runtime": 29.941,
|
| 1095 |
+
"eval_biology_samples_per_second": 16.7,
|
| 1096 |
+
"eval_biology_steps_per_second": 4.175,
|
| 1097 |
+
"step": 900
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.44,
|
| 1101 |
+
"eval_physics_entropy": 0.6614522356986999,
|
| 1102 |
+
"eval_physics_loss": 0.6613803505897522,
|
| 1103 |
+
"eval_physics_mean_token_accuracy": 0.8159068145751953,
|
| 1104 |
+
"eval_physics_num_tokens": 29789253.0,
|
| 1105 |
+
"eval_physics_runtime": 42.4043,
|
| 1106 |
+
"eval_physics_samples_per_second": 11.791,
|
| 1107 |
+
"eval_physics_steps_per_second": 2.948,
|
| 1108 |
+
"step": 900
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"entropy": 0.6264998497441411,
|
| 1112 |
+
"epoch": 1.456,
|
| 1113 |
+
"grad_norm": 15.0625,
|
| 1114 |
+
"learning_rate": 1.8180000000000002e-05,
|
| 1115 |
+
"loss": 10.0407,
|
| 1116 |
+
"mean_token_accuracy": 0.8242419868707657,
|
| 1117 |
+
"num_tokens": 30128576.0,
|
| 1118 |
+
"step": 910
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"entropy": 0.6577600754797459,
|
| 1122 |
+
"epoch": 1.472,
|
| 1123 |
+
"grad_norm": 18.875,
|
| 1124 |
+
"learning_rate": 1.8380000000000004e-05,
|
| 1125 |
+
"loss": 10.5867,
|
| 1126 |
+
"mean_token_accuracy": 0.8155373357236385,
|
| 1127 |
+
"num_tokens": 30452719.0,
|
| 1128 |
+
"step": 920
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"entropy": 0.6394334685057401,
|
| 1132 |
+
"epoch": 1.488,
|
| 1133 |
+
"grad_norm": 16.125,
|
| 1134 |
+
"learning_rate": 1.858e-05,
|
| 1135 |
+
"loss": 10.2302,
|
| 1136 |
+
"mean_token_accuracy": 0.8199584804475307,
|
| 1137 |
+
"num_tokens": 30782343.0,
|
| 1138 |
+
"step": 930
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"entropy": 0.6304329918697477,
|
| 1142 |
+
"epoch": 1.504,
|
| 1143 |
+
"grad_norm": 15.625,
|
| 1144 |
+
"learning_rate": 1.878e-05,
|
| 1145 |
+
"loss": 10.1144,
|
| 1146 |
+
"mean_token_accuracy": 0.8226039990782738,
|
| 1147 |
+
"num_tokens": 31113238.0,
|
| 1148 |
+
"step": 940
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"entropy": 0.649897889047861,
|
| 1152 |
+
"epoch": 1.52,
|
| 1153 |
+
"grad_norm": 17.25,
|
| 1154 |
+
"learning_rate": 1.898e-05,
|
| 1155 |
+
"loss": 10.4456,
|
| 1156 |
+
"mean_token_accuracy": 0.8167380917817354,
|
| 1157 |
+
"num_tokens": 31444589.0,
|
| 1158 |
+
"step": 950
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"entropy": 0.6425180744379759,
|
| 1162 |
+
"epoch": 1.536,
|
| 1163 |
+
"grad_norm": 18.375,
|
| 1164 |
+
"learning_rate": 1.918e-05,
|
| 1165 |
+
"loss": 10.3223,
|
| 1166 |
+
"mean_token_accuracy": 0.8181851245462894,
|
| 1167 |
+
"num_tokens": 31774137.0,
|
| 1168 |
+
"step": 960
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"entropy": 0.6240503186360001,
|
| 1172 |
+
"epoch": 1.552,
|
| 1173 |
+
"grad_norm": 17.375,
|
| 1174 |
+
"learning_rate": 1.938e-05,
|
| 1175 |
+
"loss": 9.9901,
|
| 1176 |
+
"mean_token_accuracy": 0.8241735894232989,
|
| 1177 |
+
"num_tokens": 32101776.0,
|
| 1178 |
+
"step": 970
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"entropy": 0.6293262420222163,
|
| 1182 |
+
"epoch": 1.568,
|
| 1183 |
+
"grad_norm": 16.625,
|
| 1184 |
+
"learning_rate": 1.9580000000000002e-05,
|
| 1185 |
+
"loss": 10.0922,
|
| 1186 |
+
"mean_token_accuracy": 0.8220919147133827,
|
| 1187 |
+
"num_tokens": 32428021.0,
|
| 1188 |
+
"step": 980
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"entropy": 0.6412162970751524,
|
| 1192 |
+
"epoch": 1.584,
|
| 1193 |
+
"grad_norm": 17.5,
|
| 1194 |
+
"learning_rate": 1.978e-05,
|
| 1195 |
+
"loss": 10.3053,
|
| 1196 |
+
"mean_token_accuracy": 0.8172495547682047,
|
| 1197 |
+
"num_tokens": 32752735.0,
|
| 1198 |
+
"step": 990
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"entropy": 0.6345738202333451,
|
| 1202 |
+
"epoch": 1.6,
|
| 1203 |
+
"grad_norm": 16.375,
|
| 1204 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 1205 |
+
"loss": 10.2193,
|
| 1206 |
+
"mean_token_accuracy": 0.8202366653829813,
|
| 1207 |
+
"num_tokens": 33075822.0,
|
| 1208 |
+
"step": 1000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 1.6,
|
| 1212 |
+
"eval_biology_entropy": 1.0772727332115173,
|
| 1213 |
+
"eval_biology_loss": 1.1990994215011597,
|
| 1214 |
+
"eval_biology_mean_token_accuracy": 0.7000869359970093,
|
| 1215 |
+
"eval_biology_num_tokens": 33075822.0,
|
| 1216 |
+
"eval_biology_runtime": 29.6208,
|
| 1217 |
+
"eval_biology_samples_per_second": 16.88,
|
| 1218 |
+
"eval_biology_steps_per_second": 4.22,
|
| 1219 |
+
"step": 1000
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 1.6,
|
| 1223 |
+
"eval_physics_entropy": 0.6503664858341217,
|
| 1224 |
+
"eval_physics_loss": 0.651718020439148,
|
| 1225 |
+
"eval_physics_mean_token_accuracy": 0.8178416604995727,
|
| 1226 |
+
"eval_physics_num_tokens": 33075822.0,
|
| 1227 |
+
"eval_physics_runtime": 42.1601,
|
| 1228 |
+
"eval_physics_samples_per_second": 11.86,
|
| 1229 |
+
"eval_physics_steps_per_second": 2.965,
|
| 1230 |
+
"step": 1000
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"entropy": 0.6182714030146599,
|
| 1234 |
+
"epoch": 1.616,
|
| 1235 |
+
"grad_norm": 17.375,
|
| 1236 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 1237 |
+
"loss": 9.8964,
|
| 1238 |
+
"mean_token_accuracy": 0.8249843519181013,
|
| 1239 |
+
"num_tokens": 33411850.0,
|
| 1240 |
+
"step": 1010
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"entropy": 0.6229340618476271,
|
| 1244 |
+
"epoch": 1.6320000000000001,
|
| 1245 |
+
"grad_norm": 16.5,
|
| 1246 |
+
"learning_rate": 1.995777777777778e-05,
|
| 1247 |
+
"loss": 10.0248,
|
| 1248 |
+
"mean_token_accuracy": 0.8223889589309692,
|
| 1249 |
+
"num_tokens": 33739456.0,
|
| 1250 |
+
"step": 1020
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"entropy": 0.6091084901243449,
|
| 1254 |
+
"epoch": 1.6480000000000001,
|
| 1255 |
+
"grad_norm": 17.5,
|
| 1256 |
+
"learning_rate": 1.9935555555555557e-05,
|
| 1257 |
+
"loss": 9.7378,
|
| 1258 |
+
"mean_token_accuracy": 0.8279588807374239,
|
| 1259 |
+
"num_tokens": 34072022.0,
|
| 1260 |
+
"step": 1030
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"entropy": 0.628602746501565,
|
| 1264 |
+
"epoch": 1.6640000000000001,
|
| 1265 |
+
"grad_norm": 18.25,
|
| 1266 |
+
"learning_rate": 1.9913333333333335e-05,
|
| 1267 |
+
"loss": 10.1041,
|
| 1268 |
+
"mean_token_accuracy": 0.8225820947438478,
|
| 1269 |
+
"num_tokens": 34398959.0,
|
| 1270 |
+
"step": 1040
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"entropy": 0.6163907073438167,
|
| 1274 |
+
"epoch": 1.6800000000000002,
|
| 1275 |
+
"grad_norm": 15.4375,
|
| 1276 |
+
"learning_rate": 1.9891111111111112e-05,
|
| 1277 |
+
"loss": 9.8895,
|
| 1278 |
+
"mean_token_accuracy": 0.824999624863267,
|
| 1279 |
+
"num_tokens": 34732284.0,
|
| 1280 |
+
"step": 1050
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"entropy": 0.6272942896932363,
|
| 1284 |
+
"epoch": 1.696,
|
| 1285 |
+
"grad_norm": 16.875,
|
| 1286 |
+
"learning_rate": 1.986888888888889e-05,
|
| 1287 |
+
"loss": 10.0732,
|
| 1288 |
+
"mean_token_accuracy": 0.822333462536335,
|
| 1289 |
+
"num_tokens": 35053712.0,
|
| 1290 |
+
"step": 1060
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"entropy": 0.6298937538638711,
|
| 1294 |
+
"epoch": 1.712,
|
| 1295 |
+
"grad_norm": 16.5,
|
| 1296 |
+
"learning_rate": 1.9846666666666668e-05,
|
| 1297 |
+
"loss": 10.1089,
|
| 1298 |
+
"mean_token_accuracy": 0.8207412119954824,
|
| 1299 |
+
"num_tokens": 35393088.0,
|
| 1300 |
+
"step": 1070
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"entropy": 0.6155203450471163,
|
| 1304 |
+
"epoch": 1.728,
|
| 1305 |
+
"grad_norm": 16.125,
|
| 1306 |
+
"learning_rate": 1.9824444444444445e-05,
|
| 1307 |
+
"loss": 9.9054,
|
| 1308 |
+
"mean_token_accuracy": 0.8255299925804138,
|
| 1309 |
+
"num_tokens": 35729007.0,
|
| 1310 |
+
"step": 1080
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"entropy": 0.6240347031503916,
|
| 1314 |
+
"epoch": 1.744,
|
| 1315 |
+
"grad_norm": 16.75,
|
| 1316 |
+
"learning_rate": 1.9802222222222226e-05,
|
| 1317 |
+
"loss": 9.9979,
|
| 1318 |
+
"mean_token_accuracy": 0.8231176160275936,
|
| 1319 |
+
"num_tokens": 36072489.0,
|
| 1320 |
+
"step": 1090
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"entropy": 0.6143232045695186,
|
| 1324 |
+
"epoch": 1.76,
|
| 1325 |
+
"grad_norm": 17.375,
|
| 1326 |
+
"learning_rate": 1.978e-05,
|
| 1327 |
+
"loss": 9.8637,
|
| 1328 |
+
"mean_token_accuracy": 0.8249252289533615,
|
| 1329 |
+
"num_tokens": 36398285.0,
|
| 1330 |
+
"step": 1100
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 1.76,
|
| 1334 |
+
"eval_biology_entropy": 1.0265369310379029,
|
| 1335 |
+
"eval_biology_loss": 1.2094172239303589,
|
| 1336 |
+
"eval_biology_mean_token_accuracy": 0.6984671940803527,
|
| 1337 |
+
"eval_biology_num_tokens": 36398285.0,
|
| 1338 |
+
"eval_biology_runtime": 29.6897,
|
| 1339 |
+
"eval_biology_samples_per_second": 16.841,
|
| 1340 |
+
"eval_biology_steps_per_second": 4.21,
|
| 1341 |
+
"step": 1100
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 1.76,
|
| 1345 |
+
"eval_physics_entropy": 0.6072226724624634,
|
| 1346 |
+
"eval_physics_loss": 0.6429793238639832,
|
| 1347 |
+
"eval_physics_mean_token_accuracy": 0.8200512890815734,
|
| 1348 |
+
"eval_physics_num_tokens": 36398285.0,
|
| 1349 |
+
"eval_physics_runtime": 42.4662,
|
| 1350 |
+
"eval_physics_samples_per_second": 11.774,
|
| 1351 |
+
"eval_physics_steps_per_second": 2.944,
|
| 1352 |
+
"step": 1100
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"entropy": 0.6318713866174221,
|
| 1356 |
+
"epoch": 1.776,
|
| 1357 |
+
"grad_norm": 15.5625,
|
| 1358 |
+
"learning_rate": 1.975777777777778e-05,
|
| 1359 |
+
"loss": 10.1889,
|
| 1360 |
+
"mean_token_accuracy": 0.8199843384325505,
|
| 1361 |
+
"num_tokens": 36730479.0,
|
| 1362 |
+
"step": 1110
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"entropy": 0.6133379619568586,
|
| 1366 |
+
"epoch": 1.792,
|
| 1367 |
+
"grad_norm": 18.125,
|
| 1368 |
+
"learning_rate": 1.9735555555555556e-05,
|
| 1369 |
+
"loss": 9.9002,
|
| 1370 |
+
"mean_token_accuracy": 0.8245061755180358,
|
| 1371 |
+
"num_tokens": 37057425.0,
|
| 1372 |
+
"step": 1120
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"entropy": 0.6075753191486001,
|
| 1376 |
+
"epoch": 1.808,
|
| 1377 |
+
"grad_norm": 15.3125,
|
| 1378 |
+
"learning_rate": 1.9713333333333337e-05,
|
| 1379 |
+
"loss": 9.7216,
|
| 1380 |
+
"mean_token_accuracy": 0.827298716455698,
|
| 1381 |
+
"num_tokens": 37395449.0,
|
| 1382 |
+
"step": 1130
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"entropy": 0.6157500127330422,
|
| 1386 |
+
"epoch": 1.8239999999999998,
|
| 1387 |
+
"grad_norm": 15.5625,
|
| 1388 |
+
"learning_rate": 1.969111111111111e-05,
|
| 1389 |
+
"loss": 9.869,
|
| 1390 |
+
"mean_token_accuracy": 0.8250419706106186,
|
| 1391 |
+
"num_tokens": 37729535.0,
|
| 1392 |
+
"step": 1140
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"entropy": 0.6038926662877202,
|
| 1396 |
+
"epoch": 1.8399999999999999,
|
| 1397 |
+
"grad_norm": 15.5625,
|
| 1398 |
+
"learning_rate": 1.9668888888888892e-05,
|
| 1399 |
+
"loss": 9.7492,
|
| 1400 |
+
"mean_token_accuracy": 0.8271189860999584,
|
| 1401 |
+
"num_tokens": 38057839.0,
|
| 1402 |
+
"step": 1150
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"entropy": 0.6130170978605747,
|
| 1406 |
+
"epoch": 1.8559999999999999,
|
| 1407 |
+
"grad_norm": 16.0,
|
| 1408 |
+
"learning_rate": 1.9646666666666666e-05,
|
| 1409 |
+
"loss": 9.7944,
|
| 1410 |
+
"mean_token_accuracy": 0.8268033184111119,
|
| 1411 |
+
"num_tokens": 38398033.0,
|
| 1412 |
+
"step": 1160
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"entropy": 0.6028442984446883,
|
| 1416 |
+
"epoch": 1.8719999999999999,
|
| 1417 |
+
"grad_norm": 16.5,
|
| 1418 |
+
"learning_rate": 1.9624444444444447e-05,
|
| 1419 |
+
"loss": 9.6695,
|
| 1420 |
+
"mean_token_accuracy": 0.8263881340622902,
|
| 1421 |
+
"num_tokens": 38729203.0,
|
| 1422 |
+
"step": 1170
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"entropy": 0.623160850442946,
|
| 1426 |
+
"epoch": 1.888,
|
| 1427 |
+
"grad_norm": 16.5,
|
| 1428 |
+
"learning_rate": 1.9602222222222225e-05,
|
| 1429 |
+
"loss": 9.9773,
|
| 1430 |
+
"mean_token_accuracy": 0.8230186630040407,
|
| 1431 |
+
"num_tokens": 39058791.0,
|
| 1432 |
+
"step": 1180
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"entropy": 0.6327232969924808,
|
| 1436 |
+
"epoch": 1.904,
|
| 1437 |
+
"grad_norm": 16.0,
|
| 1438 |
+
"learning_rate": 1.9580000000000002e-05,
|
| 1439 |
+
"loss": 10.1963,
|
| 1440 |
+
"mean_token_accuracy": 0.8210305809974671,
|
| 1441 |
+
"num_tokens": 39386170.0,
|
| 1442 |
+
"step": 1190
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"entropy": 0.6258664343506097,
|
| 1446 |
+
"epoch": 1.92,
|
| 1447 |
+
"grad_norm": 15.875,
|
| 1448 |
+
"learning_rate": 1.955777777777778e-05,
|
| 1449 |
+
"loss": 10.0632,
|
| 1450 |
+
"mean_token_accuracy": 0.8224393729120493,
|
| 1451 |
+
"num_tokens": 39710381.0,
|
| 1452 |
+
"step": 1200
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.92,
|
| 1456 |
+
"eval_biology_entropy": 1.0642137598991395,
|
| 1457 |
+
"eval_biology_loss": 1.2021751403808594,
|
| 1458 |
+
"eval_biology_mean_token_accuracy": 0.6998101119995117,
|
| 1459 |
+
"eval_biology_num_tokens": 39710381.0,
|
| 1460 |
+
"eval_biology_runtime": 29.6797,
|
| 1461 |
+
"eval_biology_samples_per_second": 16.847,
|
| 1462 |
+
"eval_biology_steps_per_second": 4.212,
|
| 1463 |
+
"step": 1200
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 1.92,
|
| 1467 |
+
"eval_physics_entropy": 0.6103473696708679,
|
| 1468 |
+
"eval_physics_loss": 0.6331072449684143,
|
| 1469 |
+
"eval_physics_mean_token_accuracy": 0.8220849981307984,
|
| 1470 |
+
"eval_physics_num_tokens": 39710381.0,
|
| 1471 |
+
"eval_physics_runtime": 42.4391,
|
| 1472 |
+
"eval_physics_samples_per_second": 11.782,
|
| 1473 |
+
"eval_physics_steps_per_second": 2.945,
|
| 1474 |
+
"step": 1200
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"entropy": 0.6147853201255202,
|
| 1478 |
+
"epoch": 1.936,
|
| 1479 |
+
"grad_norm": 15.1875,
|
| 1480 |
+
"learning_rate": 1.9535555555555557e-05,
|
| 1481 |
+
"loss": 9.8494,
|
| 1482 |
+
"mean_token_accuracy": 0.8241737693548202,
|
| 1483 |
+
"num_tokens": 40043354.0,
|
| 1484 |
+
"step": 1210
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"entropy": 0.5860691763460636,
|
| 1488 |
+
"epoch": 1.952,
|
| 1489 |
+
"grad_norm": 16.625,
|
| 1490 |
+
"learning_rate": 1.9513333333333335e-05,
|
| 1491 |
+
"loss": 9.4581,
|
| 1492 |
+
"mean_token_accuracy": 0.8311044085770846,
|
| 1493 |
+
"num_tokens": 40375790.0,
|
| 1494 |
+
"step": 1220
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"entropy": 0.6076713547110557,
|
| 1498 |
+
"epoch": 1.968,
|
| 1499 |
+
"grad_norm": 15.4375,
|
| 1500 |
+
"learning_rate": 1.9491111111111113e-05,
|
| 1501 |
+
"loss": 9.7817,
|
| 1502 |
+
"mean_token_accuracy": 0.8264643102884293,
|
| 1503 |
+
"num_tokens": 40705451.0,
|
| 1504 |
+
"step": 1230
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"entropy": 0.6071135563775897,
|
| 1508 |
+
"epoch": 1.984,
|
| 1509 |
+
"grad_norm": 16.625,
|
| 1510 |
+
"learning_rate": 1.946888888888889e-05,
|
| 1511 |
+
"loss": 9.7942,
|
| 1512 |
+
"mean_token_accuracy": 0.8252186521887779,
|
| 1513 |
+
"num_tokens": 41040313.0,
|
| 1514 |
+
"step": 1240
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"entropy": 0.6113535448908806,
|
| 1518 |
+
"epoch": 2.0,
|
| 1519 |
+
"grad_norm": 15.4375,
|
| 1520 |
+
"learning_rate": 1.9446666666666668e-05,
|
| 1521 |
+
"loss": 9.7519,
|
| 1522 |
+
"mean_token_accuracy": 0.8272297196090221,
|
| 1523 |
+
"num_tokens": 41373608.0,
|
| 1524 |
+
"step": 1250
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"entropy": 0.530447144433856,
|
| 1528 |
+
"epoch": 2.016,
|
| 1529 |
+
"grad_norm": 17.0,
|
| 1530 |
+
"learning_rate": 1.9424444444444446e-05,
|
| 1531 |
+
"loss": 8.3959,
|
| 1532 |
+
"mean_token_accuracy": 0.8450614836066961,
|
| 1533 |
+
"num_tokens": 41708226.0,
|
| 1534 |
+
"step": 1260
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"entropy": 0.5259812783449889,
|
| 1538 |
+
"epoch": 2.032,
|
| 1539 |
+
"grad_norm": 17.5,
|
| 1540 |
+
"learning_rate": 1.9402222222222223e-05,
|
| 1541 |
+
"loss": 8.4414,
|
| 1542 |
+
"mean_token_accuracy": 0.8447725884616375,
|
| 1543 |
+
"num_tokens": 42045433.0,
|
| 1544 |
+
"step": 1270
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"entropy": 0.5413103114813567,
|
| 1548 |
+
"epoch": 2.048,
|
| 1549 |
+
"grad_norm": 19.75,
|
| 1550 |
+
"learning_rate": 1.938e-05,
|
| 1551 |
+
"loss": 8.6582,
|
| 1552 |
+
"mean_token_accuracy": 0.8407626215368509,
|
| 1553 |
+
"num_tokens": 42372694.0,
|
| 1554 |
+
"step": 1280
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"entropy": 0.5283081056550145,
|
| 1558 |
+
"epoch": 2.064,
|
| 1559 |
+
"grad_norm": 17.75,
|
| 1560 |
+
"learning_rate": 1.935777777777778e-05,
|
| 1561 |
+
"loss": 8.4357,
|
| 1562 |
+
"mean_token_accuracy": 0.8448646664619446,
|
| 1563 |
+
"num_tokens": 42710240.0,
|
| 1564 |
+
"step": 1290
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"entropy": 0.5373278863728046,
|
| 1568 |
+
"epoch": 2.08,
|
| 1569 |
+
"grad_norm": 17.375,
|
| 1570 |
+
"learning_rate": 1.9335555555555556e-05,
|
| 1571 |
+
"loss": 8.6189,
|
| 1572 |
+
"mean_token_accuracy": 0.8424648057669402,
|
| 1573 |
+
"num_tokens": 43037844.0,
|
| 1574 |
+
"step": 1300
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 2.08,
|
| 1578 |
+
"eval_biology_entropy": 0.9063808312416076,
|
| 1579 |
+
"eval_biology_loss": 1.2580033540725708,
|
| 1580 |
+
"eval_biology_mean_token_accuracy": 0.6951368255615235,
|
| 1581 |
+
"eval_biology_num_tokens": 43037844.0,
|
| 1582 |
+
"eval_biology_runtime": 29.6286,
|
| 1583 |
+
"eval_biology_samples_per_second": 16.876,
|
| 1584 |
+
"eval_biology_steps_per_second": 4.219,
|
| 1585 |
+
"step": 1300
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.08,
|
| 1589 |
+
"eval_physics_entropy": 0.552986116886139,
|
| 1590 |
+
"eval_physics_loss": 0.6381449103355408,
|
| 1591 |
+
"eval_physics_mean_token_accuracy": 0.8220356435775756,
|
| 1592 |
+
"eval_physics_num_tokens": 43037844.0,
|
| 1593 |
+
"eval_physics_runtime": 42.3885,
|
| 1594 |
+
"eval_physics_samples_per_second": 11.796,
|
| 1595 |
+
"eval_physics_steps_per_second": 2.949,
|
| 1596 |
+
"step": 1300
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"entropy": 0.5337282193824648,
|
| 1600 |
+
"epoch": 2.096,
|
| 1601 |
+
"grad_norm": 16.75,
|
| 1602 |
+
"learning_rate": 1.9313333333333334e-05,
|
| 1603 |
+
"loss": 8.5123,
|
| 1604 |
+
"mean_token_accuracy": 0.8428315419703722,
|
| 1605 |
+
"num_tokens": 43368675.0,
|
| 1606 |
+
"step": 1310
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"entropy": 0.5185356924310327,
|
| 1610 |
+
"epoch": 2.112,
|
| 1611 |
+
"grad_norm": 17.0,
|
| 1612 |
+
"learning_rate": 1.9291111111111115e-05,
|
| 1613 |
+
"loss": 8.3554,
|
| 1614 |
+
"mean_token_accuracy": 0.8454546939581633,
|
| 1615 |
+
"num_tokens": 43699544.0,
|
| 1616 |
+
"step": 1320
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"entropy": 0.5311926949769259,
|
| 1620 |
+
"epoch": 2.128,
|
| 1621 |
+
"grad_norm": 17.5,
|
| 1622 |
+
"learning_rate": 1.926888888888889e-05,
|
| 1623 |
+
"loss": 8.4936,
|
| 1624 |
+
"mean_token_accuracy": 0.8429730042815209,
|
| 1625 |
+
"num_tokens": 44012632.0,
|
| 1626 |
+
"step": 1330
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"entropy": 0.5302754450589419,
|
| 1630 |
+
"epoch": 2.144,
|
| 1631 |
+
"grad_norm": 16.875,
|
| 1632 |
+
"learning_rate": 1.924666666666667e-05,
|
| 1633 |
+
"loss": 8.4927,
|
| 1634 |
+
"mean_token_accuracy": 0.8428822424262762,
|
| 1635 |
+
"num_tokens": 44349732.0,
|
| 1636 |
+
"step": 1340
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"entropy": 0.5503825809806585,
|
| 1640 |
+
"epoch": 2.16,
|
| 1641 |
+
"grad_norm": 18.0,
|
| 1642 |
+
"learning_rate": 1.9224444444444444e-05,
|
| 1643 |
+
"loss": 8.8253,
|
| 1644 |
+
"mean_token_accuracy": 0.8380614928901196,
|
| 1645 |
+
"num_tokens": 44680410.0,
|
| 1646 |
+
"step": 1350
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"entropy": 0.5267998551949858,
|
| 1650 |
+
"epoch": 2.176,
|
| 1651 |
+
"grad_norm": 16.75,
|
| 1652 |
+
"learning_rate": 1.9202222222222225e-05,
|
| 1653 |
+
"loss": 8.4749,
|
| 1654 |
+
"mean_token_accuracy": 0.8448099918663502,
|
| 1655 |
+
"num_tokens": 45017922.0,
|
| 1656 |
+
"step": 1360
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"entropy": 0.5206218842417002,
|
| 1660 |
+
"epoch": 2.192,
|
| 1661 |
+
"grad_norm": 17.0,
|
| 1662 |
+
"learning_rate": 1.918e-05,
|
| 1663 |
+
"loss": 8.3612,
|
| 1664 |
+
"mean_token_accuracy": 0.8453428644686938,
|
| 1665 |
+
"num_tokens": 45348179.0,
|
| 1666 |
+
"step": 1370
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"entropy": 0.5240026475861669,
|
| 1670 |
+
"epoch": 2.208,
|
| 1671 |
+
"grad_norm": 18.875,
|
| 1672 |
+
"learning_rate": 1.915777777777778e-05,
|
| 1673 |
+
"loss": 8.369,
|
| 1674 |
+
"mean_token_accuracy": 0.8455433510243893,
|
| 1675 |
+
"num_tokens": 45678397.0,
|
| 1676 |
+
"step": 1380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"entropy": 0.5279045928269624,
|
| 1680 |
+
"epoch": 2.224,
|
| 1681 |
+
"grad_norm": 17.0,
|
| 1682 |
+
"learning_rate": 1.9135555555555555e-05,
|
| 1683 |
+
"loss": 8.434,
|
| 1684 |
+
"mean_token_accuracy": 0.8444005087018013,
|
| 1685 |
+
"num_tokens": 46010052.0,
|
| 1686 |
+
"step": 1390
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"entropy": 0.5289217280223966,
|
| 1690 |
+
"epoch": 2.24,
|
| 1691 |
+
"grad_norm": 17.375,
|
| 1692 |
+
"learning_rate": 1.9113333333333336e-05,
|
| 1693 |
+
"loss": 8.44,
|
| 1694 |
+
"mean_token_accuracy": 0.8444702833890915,
|
| 1695 |
+
"num_tokens": 46347116.0,
|
| 1696 |
+
"step": 1400
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 2.24,
|
| 1700 |
+
"eval_biology_entropy": 0.9073377561569214,
|
| 1701 |
+
"eval_biology_loss": 1.2652989625930786,
|
| 1702 |
+
"eval_biology_mean_token_accuracy": 0.6944872965812683,
|
| 1703 |
+
"eval_biology_num_tokens": 46347116.0,
|
| 1704 |
+
"eval_biology_runtime": 29.9507,
|
| 1705 |
+
"eval_biology_samples_per_second": 16.694,
|
| 1706 |
+
"eval_biology_steps_per_second": 4.174,
|
| 1707 |
+
"step": 1400
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 2.24,
|
| 1711 |
+
"eval_physics_entropy": 0.5540535972118378,
|
| 1712 |
+
"eval_physics_loss": 0.6368283629417419,
|
| 1713 |
+
"eval_physics_mean_token_accuracy": 0.8228535957336426,
|
| 1714 |
+
"eval_physics_num_tokens": 46347116.0,
|
| 1715 |
+
"eval_physics_runtime": 42.3816,
|
| 1716 |
+
"eval_physics_samples_per_second": 11.798,
|
| 1717 |
+
"eval_physics_steps_per_second": 2.949,
|
| 1718 |
+
"step": 1400
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"entropy": 0.5402740124613047,
|
| 1722 |
+
"epoch": 2.2560000000000002,
|
| 1723 |
+
"grad_norm": 18.5,
|
| 1724 |
+
"learning_rate": 1.9091111111111113e-05,
|
| 1725 |
+
"loss": 8.7016,
|
| 1726 |
+
"mean_token_accuracy": 0.8409985613077879,
|
| 1727 |
+
"num_tokens": 46673074.0,
|
| 1728 |
+
"step": 1410
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"entropy": 0.5238024350255728,
|
| 1732 |
+
"epoch": 2.2720000000000002,
|
| 1733 |
+
"grad_norm": 18.75,
|
| 1734 |
+
"learning_rate": 1.906888888888889e-05,
|
| 1735 |
+
"loss": 8.3724,
|
| 1736 |
+
"mean_token_accuracy": 0.8452999964356422,
|
| 1737 |
+
"num_tokens": 47009386.0,
|
| 1738 |
+
"step": 1420
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"entropy": 0.5359452277421951,
|
| 1742 |
+
"epoch": 2.288,
|
| 1743 |
+
"grad_norm": 17.625,
|
| 1744 |
+
"learning_rate": 1.904666666666667e-05,
|
| 1745 |
+
"loss": 8.6322,
|
| 1746 |
+
"mean_token_accuracy": 0.8418019615113735,
|
| 1747 |
+
"num_tokens": 47344231.0,
|
| 1748 |
+
"step": 1430
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"entropy": 0.5372845523059369,
|
| 1752 |
+
"epoch": 2.304,
|
| 1753 |
+
"grad_norm": 19.625,
|
| 1754 |
+
"learning_rate": 1.9024444444444446e-05,
|
| 1755 |
+
"loss": 8.5597,
|
| 1756 |
+
"mean_token_accuracy": 0.8422956768423319,
|
| 1757 |
+
"num_tokens": 47658392.0,
|
| 1758 |
+
"step": 1440
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"entropy": 0.5315724445506931,
|
| 1762 |
+
"epoch": 2.32,
|
| 1763 |
+
"grad_norm": 16.125,
|
| 1764 |
+
"learning_rate": 1.9002222222222224e-05,
|
| 1765 |
+
"loss": 8.5227,
|
| 1766 |
+
"mean_token_accuracy": 0.8427653182297945,
|
| 1767 |
+
"num_tokens": 47995015.0,
|
| 1768 |
+
"step": 1450
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"entropy": 0.5130708247423172,
|
| 1772 |
+
"epoch": 2.336,
|
| 1773 |
+
"grad_norm": 18.125,
|
| 1774 |
+
"learning_rate": 1.898e-05,
|
| 1775 |
+
"loss": 8.2514,
|
| 1776 |
+
"mean_token_accuracy": 0.8464119978249073,
|
| 1777 |
+
"num_tokens": 48336708.0,
|
| 1778 |
+
"step": 1460
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"entropy": 0.5171284712851048,
|
| 1782 |
+
"epoch": 2.352,
|
| 1783 |
+
"grad_norm": 17.625,
|
| 1784 |
+
"learning_rate": 1.895777777777778e-05,
|
| 1785 |
+
"loss": 8.2526,
|
| 1786 |
+
"mean_token_accuracy": 0.8467188984155655,
|
| 1787 |
+
"num_tokens": 48665311.0,
|
| 1788 |
+
"step": 1470
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"entropy": 0.5282838650047779,
|
| 1792 |
+
"epoch": 2.368,
|
| 1793 |
+
"grad_norm": 17.25,
|
| 1794 |
+
"learning_rate": 1.8935555555555556e-05,
|
| 1795 |
+
"loss": 8.4841,
|
| 1796 |
+
"mean_token_accuracy": 0.8436643645167351,
|
| 1797 |
+
"num_tokens": 49005334.0,
|
| 1798 |
+
"step": 1480
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"entropy": 0.534802301414311,
|
| 1802 |
+
"epoch": 2.384,
|
| 1803 |
+
"grad_norm": 19.375,
|
| 1804 |
+
"learning_rate": 1.8913333333333334e-05,
|
| 1805 |
+
"loss": 8.5751,
|
| 1806 |
+
"mean_token_accuracy": 0.8415927398949862,
|
| 1807 |
+
"num_tokens": 49331936.0,
|
| 1808 |
+
"step": 1490
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"entropy": 0.5311837416142225,
|
| 1812 |
+
"epoch": 2.4,
|
| 1813 |
+
"grad_norm": 18.625,
|
| 1814 |
+
"learning_rate": 1.8891111111111115e-05,
|
| 1815 |
+
"loss": 8.5063,
|
| 1816 |
+
"mean_token_accuracy": 0.8436322450637818,
|
| 1817 |
+
"num_tokens": 49671048.0,
|
| 1818 |
+
"step": 1500
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 2.4,
|
| 1822 |
+
"eval_biology_entropy": 0.8830818734169006,
|
| 1823 |
+
"eval_biology_loss": 1.2704880237579346,
|
| 1824 |
+
"eval_biology_mean_token_accuracy": 0.6929694819450378,
|
| 1825 |
+
"eval_biology_num_tokens": 49671048.0,
|
| 1826 |
+
"eval_biology_runtime": 29.626,
|
| 1827 |
+
"eval_biology_samples_per_second": 16.877,
|
| 1828 |
+
"eval_biology_steps_per_second": 4.219,
|
| 1829 |
+
"step": 1500
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 2.4,
|
| 1833 |
+
"eval_physics_entropy": 0.540393363237381,
|
| 1834 |
+
"eval_physics_loss": 0.6307452321052551,
|
| 1835 |
+
"eval_physics_mean_token_accuracy": 0.8239317517280579,
|
| 1836 |
+
"eval_physics_num_tokens": 49671048.0,
|
| 1837 |
+
"eval_physics_runtime": 42.2546,
|
| 1838 |
+
"eval_physics_samples_per_second": 11.833,
|
| 1839 |
+
"eval_physics_steps_per_second": 2.958,
|
| 1840 |
+
"step": 1500
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"entropy": 0.5313838511705399,
|
| 1844 |
+
"epoch": 2.416,
|
| 1845 |
+
"grad_norm": 17.25,
|
| 1846 |
+
"learning_rate": 1.886888888888889e-05,
|
| 1847 |
+
"loss": 8.5314,
|
| 1848 |
+
"mean_token_accuracy": 0.8434251178056001,
|
| 1849 |
+
"num_tokens": 50018182.0,
|
| 1850 |
+
"step": 1510
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"entropy": 0.5314315013587475,
|
| 1854 |
+
"epoch": 2.432,
|
| 1855 |
+
"grad_norm": 18.875,
|
| 1856 |
+
"learning_rate": 1.884666666666667e-05,
|
| 1857 |
+
"loss": 8.5551,
|
| 1858 |
+
"mean_token_accuracy": 0.8438528429716825,
|
| 1859 |
+
"num_tokens": 50341187.0,
|
| 1860 |
+
"step": 1520
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"entropy": 0.5497623635455966,
|
| 1864 |
+
"epoch": 2.448,
|
| 1865 |
+
"grad_norm": 17.875,
|
| 1866 |
+
"learning_rate": 1.8824444444444445e-05,
|
| 1867 |
+
"loss": 8.769,
|
| 1868 |
+
"mean_token_accuracy": 0.839348716661334,
|
| 1869 |
+
"num_tokens": 50667019.0,
|
| 1870 |
+
"step": 1530
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"entropy": 0.5337263094261289,
|
| 1874 |
+
"epoch": 2.464,
|
| 1875 |
+
"grad_norm": 18.125,
|
| 1876 |
+
"learning_rate": 1.8802222222222226e-05,
|
| 1877 |
+
"loss": 8.5788,
|
| 1878 |
+
"mean_token_accuracy": 0.8423588499426842,
|
| 1879 |
+
"num_tokens": 50991113.0,
|
| 1880 |
+
"step": 1540
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"entropy": 0.533509024977684,
|
| 1884 |
+
"epoch": 2.48,
|
| 1885 |
+
"grad_norm": 18.25,
|
| 1886 |
+
"learning_rate": 1.878e-05,
|
| 1887 |
+
"loss": 8.5669,
|
| 1888 |
+
"mean_token_accuracy": 0.8423088423907756,
|
| 1889 |
+
"num_tokens": 51319624.0,
|
| 1890 |
+
"step": 1550
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"entropy": 0.5355414936318994,
|
| 1894 |
+
"epoch": 2.496,
|
| 1895 |
+
"grad_norm": 18.0,
|
| 1896 |
+
"learning_rate": 1.875777777777778e-05,
|
| 1897 |
+
"loss": 8.5578,
|
| 1898 |
+
"mean_token_accuracy": 0.8425650514662266,
|
| 1899 |
+
"num_tokens": 51654633.0,
|
| 1900 |
+
"step": 1560
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"entropy": 0.5151204424910247,
|
| 1904 |
+
"epoch": 2.512,
|
| 1905 |
+
"grad_norm": 18.75,
|
| 1906 |
+
"learning_rate": 1.873555555555556e-05,
|
| 1907 |
+
"loss": 8.3018,
|
| 1908 |
+
"mean_token_accuracy": 0.8466352391988039,
|
| 1909 |
+
"num_tokens": 51980051.0,
|
| 1910 |
+
"step": 1570
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"entropy": 0.5199752731248737,
|
| 1914 |
+
"epoch": 2.528,
|
| 1915 |
+
"grad_norm": 16.375,
|
| 1916 |
+
"learning_rate": 1.8713333333333336e-05,
|
| 1917 |
+
"loss": 8.3413,
|
| 1918 |
+
"mean_token_accuracy": 0.8458983488380909,
|
| 1919 |
+
"num_tokens": 52316109.0,
|
| 1920 |
+
"step": 1580
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"entropy": 0.5363344758749008,
|
| 1924 |
+
"epoch": 2.544,
|
| 1925 |
+
"grad_norm": 19.0,
|
| 1926 |
+
"learning_rate": 1.8691111111111114e-05,
|
| 1927 |
+
"loss": 8.5941,
|
| 1928 |
+
"mean_token_accuracy": 0.8419183246791363,
|
| 1929 |
+
"num_tokens": 52634528.0,
|
| 1930 |
+
"step": 1590
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"entropy": 0.5285956017673016,
|
| 1934 |
+
"epoch": 2.56,
|
| 1935 |
+
"grad_norm": 17.625,
|
| 1936 |
+
"learning_rate": 1.866888888888889e-05,
|
| 1937 |
+
"loss": 8.433,
|
| 1938 |
+
"mean_token_accuracy": 0.8443803623318672,
|
| 1939 |
+
"num_tokens": 52964035.0,
|
| 1940 |
+
"step": 1600
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 2.56,
|
| 1944 |
+
"eval_biology_entropy": 0.9072599239349365,
|
| 1945 |
+
"eval_biology_loss": 1.264472246170044,
|
| 1946 |
+
"eval_biology_mean_token_accuracy": 0.6935372724533081,
|
| 1947 |
+
"eval_biology_num_tokens": 52964035.0,
|
| 1948 |
+
"eval_biology_runtime": 29.6445,
|
| 1949 |
+
"eval_biology_samples_per_second": 16.867,
|
| 1950 |
+
"eval_biology_steps_per_second": 4.217,
|
| 1951 |
+
"step": 1600
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 2.56,
|
| 1955 |
+
"eval_physics_entropy": 0.5502645847797394,
|
| 1956 |
+
"eval_physics_loss": 0.6250398755073547,
|
| 1957 |
+
"eval_physics_mean_token_accuracy": 0.8245352702140808,
|
| 1958 |
+
"eval_physics_num_tokens": 52964035.0,
|
| 1959 |
+
"eval_physics_runtime": 42.3903,
|
| 1960 |
+
"eval_physics_samples_per_second": 11.795,
|
| 1961 |
+
"eval_physics_steps_per_second": 2.949,
|
| 1962 |
+
"step": 1600
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"entropy": 0.5171479491516948,
|
| 1966 |
+
"epoch": 2.576,
|
| 1967 |
+
"grad_norm": 18.125,
|
| 1968 |
+
"learning_rate": 1.864666666666667e-05,
|
| 1969 |
+
"loss": 8.3356,
|
| 1970 |
+
"mean_token_accuracy": 0.8456523548811674,
|
| 1971 |
+
"num_tokens": 53293345.0,
|
| 1972 |
+
"step": 1610
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"entropy": 0.5298814970999957,
|
| 1976 |
+
"epoch": 2.592,
|
| 1977 |
+
"grad_norm": 16.125,
|
| 1978 |
+
"learning_rate": 1.8624444444444446e-05,
|
| 1979 |
+
"loss": 8.4943,
|
| 1980 |
+
"mean_token_accuracy": 0.8433435622602701,
|
| 1981 |
+
"num_tokens": 53620896.0,
|
| 1982 |
+
"step": 1620
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"entropy": 0.5211699862033129,
|
| 1986 |
+
"epoch": 2.608,
|
| 1987 |
+
"grad_norm": 16.25,
|
| 1988 |
+
"learning_rate": 1.8602222222222224e-05,
|
| 1989 |
+
"loss": 8.3802,
|
| 1990 |
+
"mean_token_accuracy": 0.8462613090872765,
|
| 1991 |
+
"num_tokens": 53950989.0,
|
| 1992 |
+
"step": 1630
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"entropy": 0.5339049361646175,
|
| 1996 |
+
"epoch": 2.624,
|
| 1997 |
+
"grad_norm": 17.0,
|
| 1998 |
+
"learning_rate": 1.858e-05,
|
| 1999 |
+
"loss": 8.5302,
|
| 2000 |
+
"mean_token_accuracy": 0.8435598284006118,
|
| 2001 |
+
"num_tokens": 54282732.0,
|
| 2002 |
+
"step": 1640
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"entropy": 0.5256838278844953,
|
| 2006 |
+
"epoch": 2.64,
|
| 2007 |
+
"grad_norm": 17.75,
|
| 2008 |
+
"learning_rate": 1.855777777777778e-05,
|
| 2009 |
+
"loss": 8.473,
|
| 2010 |
+
"mean_token_accuracy": 0.8441216733306647,
|
| 2011 |
+
"num_tokens": 54609748.0,
|
| 2012 |
+
"step": 1650
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"entropy": 0.5526315724477172,
|
| 2016 |
+
"epoch": 2.656,
|
| 2017 |
+
"grad_norm": 19.5,
|
| 2018 |
+
"learning_rate": 1.8535555555555557e-05,
|
| 2019 |
+
"loss": 8.8463,
|
| 2020 |
+
"mean_token_accuracy": 0.8381063390523196,
|
| 2021 |
+
"num_tokens": 54953583.0,
|
| 2022 |
+
"step": 1660
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"entropy": 0.5346001913771034,
|
| 2026 |
+
"epoch": 2.672,
|
| 2027 |
+
"grad_norm": 17.875,
|
| 2028 |
+
"learning_rate": 1.8513333333333335e-05,
|
| 2029 |
+
"loss": 8.5872,
|
| 2030 |
+
"mean_token_accuracy": 0.8416836660355329,
|
| 2031 |
+
"num_tokens": 55281945.0,
|
| 2032 |
+
"step": 1670
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"entropy": 0.507072365656495,
|
| 2036 |
+
"epoch": 2.6879999999999997,
|
| 2037 |
+
"grad_norm": 17.0,
|
| 2038 |
+
"learning_rate": 1.8491111111111112e-05,
|
| 2039 |
+
"loss": 8.1723,
|
| 2040 |
+
"mean_token_accuracy": 0.848843240365386,
|
| 2041 |
+
"num_tokens": 55612391.0,
|
| 2042 |
+
"step": 1680
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"entropy": 0.539692627824843,
|
| 2046 |
+
"epoch": 2.7039999999999997,
|
| 2047 |
+
"grad_norm": 17.25,
|
| 2048 |
+
"learning_rate": 1.846888888888889e-05,
|
| 2049 |
+
"loss": 8.6013,
|
| 2050 |
+
"mean_token_accuracy": 0.8423799473792315,
|
| 2051 |
+
"num_tokens": 55949791.0,
|
| 2052 |
+
"step": 1690
|
| 2053 |
+
},
|
| 2054 |
+
{
|
| 2055 |
+
"entropy": 0.5213707761839033,
|
| 2056 |
+
"epoch": 2.7199999999999998,
|
| 2057 |
+
"grad_norm": 20.875,
|
| 2058 |
+
"learning_rate": 1.8446666666666667e-05,
|
| 2059 |
+
"loss": 8.3781,
|
| 2060 |
+
"mean_token_accuracy": 0.8448753044009208,
|
| 2061 |
+
"num_tokens": 56282204.0,
|
| 2062 |
+
"step": 1700
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 2.7199999999999998,
|
| 2066 |
+
"eval_biology_entropy": 0.8998830423355103,
|
| 2067 |
+
"eval_biology_loss": 1.269989013671875,
|
| 2068 |
+
"eval_biology_mean_token_accuracy": 0.6936762638092041,
|
| 2069 |
+
"eval_biology_num_tokens": 56282204.0,
|
| 2070 |
+
"eval_biology_runtime": 29.629,
|
| 2071 |
+
"eval_biology_samples_per_second": 16.875,
|
| 2072 |
+
"eval_biology_steps_per_second": 4.219,
|
| 2073 |
+
"step": 1700
|
| 2074 |
+
},
|
| 2075 |
+
{
|
| 2076 |
+
"epoch": 2.7199999999999998,
|
| 2077 |
+
"eval_physics_entropy": 0.5432275445461273,
|
| 2078 |
+
"eval_physics_loss": 0.6207777261734009,
|
| 2079 |
+
"eval_physics_mean_token_accuracy": 0.8257771005630493,
|
| 2080 |
+
"eval_physics_num_tokens": 56282204.0,
|
| 2081 |
+
"eval_physics_runtime": 42.3855,
|
| 2082 |
+
"eval_physics_samples_per_second": 11.796,
|
| 2083 |
+
"eval_physics_steps_per_second": 2.949,
|
| 2084 |
+
"step": 1700
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"entropy": 0.5400118172168732,
|
| 2088 |
+
"epoch": 2.7359999999999998,
|
| 2089 |
+
"grad_norm": 18.625,
|
| 2090 |
+
"learning_rate": 1.842444444444445e-05,
|
| 2091 |
+
"loss": 8.6625,
|
| 2092 |
+
"mean_token_accuracy": 0.8404322098940611,
|
| 2093 |
+
"num_tokens": 56605478.0,
|
| 2094 |
+
"step": 1710
|
| 2095 |
+
},
|
| 2096 |
+
{
|
| 2097 |
+
"entropy": 0.5394736461341381,
|
| 2098 |
+
"epoch": 2.752,
|
| 2099 |
+
"grad_norm": 17.625,
|
| 2100 |
+
"learning_rate": 1.8402222222222223e-05,
|
| 2101 |
+
"loss": 8.6358,
|
| 2102 |
+
"mean_token_accuracy": 0.8410456858575344,
|
| 2103 |
+
"num_tokens": 56935874.0,
|
| 2104 |
+
"step": 1720
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"entropy": 0.5178206816315651,
|
| 2108 |
+
"epoch": 2.768,
|
| 2109 |
+
"grad_norm": 17.875,
|
| 2110 |
+
"learning_rate": 1.8380000000000004e-05,
|
| 2111 |
+
"loss": 8.337,
|
| 2112 |
+
"mean_token_accuracy": 0.8459576744586229,
|
| 2113 |
+
"num_tokens": 57270552.0,
|
| 2114 |
+
"step": 1730
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"entropy": 0.5335245341062546,
|
| 2118 |
+
"epoch": 2.784,
|
| 2119 |
+
"grad_norm": 18.125,
|
| 2120 |
+
"learning_rate": 1.8357777777777778e-05,
|
| 2121 |
+
"loss": 8.5249,
|
| 2122 |
+
"mean_token_accuracy": 0.8427888829261064,
|
| 2123 |
+
"num_tokens": 57602968.0,
|
| 2124 |
+
"step": 1740
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"entropy": 0.5528621880337596,
|
| 2128 |
+
"epoch": 2.8,
|
| 2129 |
+
"grad_norm": 18.25,
|
| 2130 |
+
"learning_rate": 1.833555555555556e-05,
|
| 2131 |
+
"loss": 8.8497,
|
| 2132 |
+
"mean_token_accuracy": 0.8369761358946561,
|
| 2133 |
+
"num_tokens": 57925886.0,
|
| 2134 |
+
"step": 1750
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"entropy": 0.5134829036891461,
|
| 2138 |
+
"epoch": 2.816,
|
| 2139 |
+
"grad_norm": 16.75,
|
| 2140 |
+
"learning_rate": 1.8313333333333333e-05,
|
| 2141 |
+
"loss": 8.1668,
|
| 2142 |
+
"mean_token_accuracy": 0.848399018868804,
|
| 2143 |
+
"num_tokens": 58262960.0,
|
| 2144 |
+
"step": 1760
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"entropy": 0.5281499922275543,
|
| 2148 |
+
"epoch": 2.832,
|
| 2149 |
+
"grad_norm": 17.75,
|
| 2150 |
+
"learning_rate": 1.8291111111111114e-05,
|
| 2151 |
+
"loss": 8.5747,
|
| 2152 |
+
"mean_token_accuracy": 0.8418189071118831,
|
| 2153 |
+
"num_tokens": 58586324.0,
|
| 2154 |
+
"step": 1770
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"entropy": 0.5403068274259567,
|
| 2158 |
+
"epoch": 2.848,
|
| 2159 |
+
"grad_norm": 17.125,
|
| 2160 |
+
"learning_rate": 1.8268888888888888e-05,
|
| 2161 |
+
"loss": 8.6062,
|
| 2162 |
+
"mean_token_accuracy": 0.8411106329411269,
|
| 2163 |
+
"num_tokens": 58912949.0,
|
| 2164 |
+
"step": 1780
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"entropy": 0.5275976810604334,
|
| 2168 |
+
"epoch": 2.864,
|
| 2169 |
+
"grad_norm": 16.125,
|
| 2170 |
+
"learning_rate": 1.824666666666667e-05,
|
| 2171 |
+
"loss": 8.4806,
|
| 2172 |
+
"mean_token_accuracy": 0.844698841124773,
|
| 2173 |
+
"num_tokens": 59252494.0,
|
| 2174 |
+
"step": 1790
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"entropy": 0.5257823524996639,
|
| 2178 |
+
"epoch": 2.88,
|
| 2179 |
+
"grad_norm": 18.875,
|
| 2180 |
+
"learning_rate": 1.8224444444444447e-05,
|
| 2181 |
+
"loss": 8.4506,
|
| 2182 |
+
"mean_token_accuracy": 0.8446649070829153,
|
| 2183 |
+
"num_tokens": 59574951.0,
|
| 2184 |
+
"step": 1800
|
| 2185 |
+
},
|
| 2186 |
+
{
|
| 2187 |
+
"epoch": 2.88,
|
| 2188 |
+
"eval_biology_entropy": 0.91778861618042,
|
| 2189 |
+
"eval_biology_loss": 1.2650281190872192,
|
| 2190 |
+
"eval_biology_mean_token_accuracy": 0.69282026720047,
|
| 2191 |
+
"eval_biology_num_tokens": 59574951.0,
|
| 2192 |
+
"eval_biology_runtime": 29.6226,
|
| 2193 |
+
"eval_biology_samples_per_second": 16.879,
|
| 2194 |
+
"eval_biology_steps_per_second": 4.22,
|
| 2195 |
+
"step": 1800
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 2.88,
|
| 2199 |
+
"eval_physics_entropy": 0.5498799192905426,
|
| 2200 |
+
"eval_physics_loss": 0.6160721182823181,
|
| 2201 |
+
"eval_physics_mean_token_accuracy": 0.8270136556625366,
|
| 2202 |
+
"eval_physics_num_tokens": 59574951.0,
|
| 2203 |
+
"eval_physics_runtime": 42.3762,
|
| 2204 |
+
"eval_physics_samples_per_second": 11.799,
|
| 2205 |
+
"eval_physics_steps_per_second": 2.95,
|
| 2206 |
+
"step": 1800
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"entropy": 0.5232849149033427,
|
| 2210 |
+
"epoch": 2.896,
|
| 2211 |
+
"grad_norm": 17.625,
|
| 2212 |
+
"learning_rate": 1.8202222222222225e-05,
|
| 2213 |
+
"loss": 8.4238,
|
| 2214 |
+
"mean_token_accuracy": 0.8446525242179632,
|
| 2215 |
+
"num_tokens": 59903274.0,
|
| 2216 |
+
"step": 1810
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"entropy": 0.5211154764518142,
|
| 2220 |
+
"epoch": 2.912,
|
| 2221 |
+
"grad_norm": 17.375,
|
| 2222 |
+
"learning_rate": 1.8180000000000002e-05,
|
| 2223 |
+
"loss": 8.3389,
|
| 2224 |
+
"mean_token_accuracy": 0.8464620877057314,
|
| 2225 |
+
"num_tokens": 60234203.0,
|
| 2226 |
+
"step": 1820
|
| 2227 |
+
},
|
| 2228 |
+
{
|
| 2229 |
+
"entropy": 0.5203663635998964,
|
| 2230 |
+
"epoch": 2.928,
|
| 2231 |
+
"grad_norm": 17.75,
|
| 2232 |
+
"learning_rate": 1.815777777777778e-05,
|
| 2233 |
+
"loss": 8.3211,
|
| 2234 |
+
"mean_token_accuracy": 0.8467351488769055,
|
| 2235 |
+
"num_tokens": 60559123.0,
|
| 2236 |
+
"step": 1830
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"entropy": 0.5172538913786411,
|
| 2240 |
+
"epoch": 2.944,
|
| 2241 |
+
"grad_norm": 19.375,
|
| 2242 |
+
"learning_rate": 1.8135555555555557e-05,
|
| 2243 |
+
"loss": 8.3406,
|
| 2244 |
+
"mean_token_accuracy": 0.8462818019092083,
|
| 2245 |
+
"num_tokens": 60893423.0,
|
| 2246 |
+
"step": 1840
|
| 2247 |
+
},
|
| 2248 |
+
{
|
| 2249 |
+
"entropy": 0.5285945516079664,
|
| 2250 |
+
"epoch": 2.96,
|
| 2251 |
+
"grad_norm": 17.125,
|
| 2252 |
+
"learning_rate": 1.8113333333333335e-05,
|
| 2253 |
+
"loss": 8.4533,
|
| 2254 |
+
"mean_token_accuracy": 0.8438907958567142,
|
| 2255 |
+
"num_tokens": 61228692.0,
|
| 2256 |
+
"step": 1850
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"entropy": 0.5289302395656705,
|
| 2260 |
+
"epoch": 2.976,
|
| 2261 |
+
"grad_norm": 17.5,
|
| 2262 |
+
"learning_rate": 1.8091111111111113e-05,
|
| 2263 |
+
"loss": 8.4515,
|
| 2264 |
+
"mean_token_accuracy": 0.8438733883202076,
|
| 2265 |
+
"num_tokens": 61560852.0,
|
| 2266 |
+
"step": 1860
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"entropy": 0.512842470407486,
|
| 2270 |
+
"epoch": 2.992,
|
| 2271 |
+
"grad_norm": 17.625,
|
| 2272 |
+
"learning_rate": 1.806888888888889e-05,
|
| 2273 |
+
"loss": 8.2149,
|
| 2274 |
+
"mean_token_accuracy": 0.8474430676549674,
|
| 2275 |
+
"num_tokens": 61898951.0,
|
| 2276 |
+
"step": 1870
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"entropy": 0.4715210357680917,
|
| 2280 |
+
"epoch": 3.008,
|
| 2281 |
+
"grad_norm": 23.625,
|
| 2282 |
+
"learning_rate": 1.8046666666666668e-05,
|
| 2283 |
+
"loss": 7.2166,
|
| 2284 |
+
"mean_token_accuracy": 0.8635042041540146,
|
| 2285 |
+
"num_tokens": 62223875.0,
|
| 2286 |
+
"step": 1880
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"entropy": 0.37948312778025867,
|
| 2290 |
+
"epoch": 3.024,
|
| 2291 |
+
"grad_norm": 23.5,
|
| 2292 |
+
"learning_rate": 1.8024444444444445e-05,
|
| 2293 |
+
"loss": 6.161,
|
| 2294 |
+
"mean_token_accuracy": 0.880422555282712,
|
| 2295 |
+
"num_tokens": 62553166.0,
|
| 2296 |
+
"step": 1890
|
| 2297 |
+
},
|
| 2298 |
+
{
|
| 2299 |
+
"entropy": 0.3847277104854584,
|
| 2300 |
+
"epoch": 3.04,
|
| 2301 |
+
"grad_norm": 24.0,
|
| 2302 |
+
"learning_rate": 1.8002222222222223e-05,
|
| 2303 |
+
"loss": 6.1092,
|
| 2304 |
+
"mean_token_accuracy": 0.8813776560127735,
|
| 2305 |
+
"num_tokens": 62893113.0,
|
| 2306 |
+
"step": 1900
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 3.04,
|
| 2310 |
+
"eval_biology_entropy": 0.6869136800765991,
|
| 2311 |
+
"eval_biology_loss": 1.5112156867980957,
|
| 2312 |
+
"eval_biology_mean_token_accuracy": 0.6802826480865478,
|
| 2313 |
+
"eval_biology_num_tokens": 62893113.0,
|
| 2314 |
+
"eval_biology_runtime": 29.5983,
|
| 2315 |
+
"eval_biology_samples_per_second": 16.893,
|
| 2316 |
+
"eval_biology_steps_per_second": 4.223,
|
| 2317 |
+
"step": 1900
|
| 2318 |
+
},
|
| 2319 |
+
{
|
| 2320 |
+
"epoch": 3.04,
|
| 2321 |
+
"eval_physics_entropy": 0.42201741313934327,
|
| 2322 |
+
"eval_physics_loss": 0.6906136870384216,
|
| 2323 |
+
"eval_physics_mean_token_accuracy": 0.8213669605255127,
|
| 2324 |
+
"eval_physics_num_tokens": 62893113.0,
|
| 2325 |
+
"eval_physics_runtime": 42.6547,
|
| 2326 |
+
"eval_physics_samples_per_second": 11.722,
|
| 2327 |
+
"eval_physics_steps_per_second": 2.931,
|
| 2328 |
+
"step": 1900
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"entropy": 0.37959762578830125,
|
| 2332 |
+
"epoch": 3.056,
|
| 2333 |
+
"grad_norm": 23.25,
|
| 2334 |
+
"learning_rate": 1.798e-05,
|
| 2335 |
+
"loss": 6.0285,
|
| 2336 |
+
"mean_token_accuracy": 0.8816115573048592,
|
| 2337 |
+
"num_tokens": 63218088.0,
|
| 2338 |
+
"step": 1910
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"entropy": 0.3821444820612669,
|
| 2342 |
+
"epoch": 3.072,
|
| 2343 |
+
"grad_norm": 24.875,
|
| 2344 |
+
"learning_rate": 1.7957777777777778e-05,
|
| 2345 |
+
"loss": 6.0732,
|
| 2346 |
+
"mean_token_accuracy": 0.8815719917416572,
|
| 2347 |
+
"num_tokens": 63551201.0,
|
| 2348 |
+
"step": 1920
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"entropy": 0.39367833686992526,
|
| 2352 |
+
"epoch": 3.088,
|
| 2353 |
+
"grad_norm": 24.875,
|
| 2354 |
+
"learning_rate": 1.7935555555555556e-05,
|
| 2355 |
+
"loss": 6.2704,
|
| 2356 |
+
"mean_token_accuracy": 0.8780518177896738,
|
| 2357 |
+
"num_tokens": 63876362.0,
|
| 2358 |
+
"step": 1930
|
| 2359 |
+
},
|
| 2360 |
+
{
|
| 2361 |
+
"entropy": 0.3746034217998385,
|
| 2362 |
+
"epoch": 3.104,
|
| 2363 |
+
"grad_norm": 23.75,
|
| 2364 |
+
"learning_rate": 1.7913333333333337e-05,
|
| 2365 |
+
"loss": 5.9319,
|
| 2366 |
+
"mean_token_accuracy": 0.8839064195752144,
|
| 2367 |
+
"num_tokens": 64208171.0,
|
| 2368 |
+
"step": 1940
|
| 2369 |
+
},
|
| 2370 |
+
{
|
| 2371 |
+
"entropy": 0.37151786470785736,
|
| 2372 |
+
"epoch": 3.12,
|
| 2373 |
+
"grad_norm": 23.875,
|
| 2374 |
+
"learning_rate": 1.789111111111111e-05,
|
| 2375 |
+
"loss": 5.9241,
|
| 2376 |
+
"mean_token_accuracy": 0.884624307230115,
|
| 2377 |
+
"num_tokens": 64544493.0,
|
| 2378 |
+
"step": 1950
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"entropy": 0.39232265017926693,
|
| 2382 |
+
"epoch": 3.136,
|
| 2383 |
+
"grad_norm": 23.625,
|
| 2384 |
+
"learning_rate": 1.7868888888888892e-05,
|
| 2385 |
+
"loss": 6.2659,
|
| 2386 |
+
"mean_token_accuracy": 0.87808175124228,
|
| 2387 |
+
"num_tokens": 64872488.0,
|
| 2388 |
+
"step": 1960
|
| 2389 |
+
},
|
| 2390 |
+
{
|
| 2391 |
+
"entropy": 0.37443468971177934,
|
| 2392 |
+
"epoch": 3.152,
|
| 2393 |
+
"grad_norm": 24.0,
|
| 2394 |
+
"learning_rate": 1.7846666666666666e-05,
|
| 2395 |
+
"loss": 5.9417,
|
| 2396 |
+
"mean_token_accuracy": 0.8836083229631185,
|
| 2397 |
+
"num_tokens": 65201124.0,
|
| 2398 |
+
"step": 1970
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"entropy": 0.385872579459101,
|
| 2402 |
+
"epoch": 3.168,
|
| 2403 |
+
"grad_norm": 23.0,
|
| 2404 |
+
"learning_rate": 1.7824444444444447e-05,
|
| 2405 |
+
"loss": 6.1646,
|
| 2406 |
+
"mean_token_accuracy": 0.8805806953459978,
|
| 2407 |
+
"num_tokens": 65525868.0,
|
| 2408 |
+
"step": 1980
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"entropy": 0.37636062121018765,
|
| 2412 |
+
"epoch": 3.184,
|
| 2413 |
+
"grad_norm": 24.0,
|
| 2414 |
+
"learning_rate": 1.780222222222222e-05,
|
| 2415 |
+
"loss": 5.9906,
|
| 2416 |
+
"mean_token_accuracy": 0.883328914642334,
|
| 2417 |
+
"num_tokens": 65861603.0,
|
| 2418 |
+
"step": 1990
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"entropy": 0.3804218841716647,
|
| 2422 |
+
"epoch": 3.2,
|
| 2423 |
+
"grad_norm": 23.75,
|
| 2424 |
+
"learning_rate": 1.7780000000000003e-05,
|
| 2425 |
+
"loss": 6.1031,
|
| 2426 |
+
"mean_token_accuracy": 0.8820050243288279,
|
| 2427 |
+
"num_tokens": 66193675.0,
|
| 2428 |
+
"step": 2000
|
| 2429 |
+
},
|
| 2430 |
+
{
|
| 2431 |
+
"epoch": 3.2,
|
| 2432 |
+
"eval_biology_entropy": 0.6348137440681457,
|
| 2433 |
+
"eval_biology_loss": 1.5773308277130127,
|
| 2434 |
+
"eval_biology_mean_token_accuracy": 0.6818766250610352,
|
| 2435 |
+
"eval_biology_num_tokens": 66193675.0,
|
| 2436 |
+
"eval_biology_runtime": 29.6404,
|
| 2437 |
+
"eval_biology_samples_per_second": 16.869,
|
| 2438 |
+
"eval_biology_steps_per_second": 4.217,
|
| 2439 |
+
"step": 2000
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 3.2,
|
| 2443 |
+
"eval_physics_entropy": 0.4191882491111755,
|
| 2444 |
+
"eval_physics_loss": 0.690969705581665,
|
| 2445 |
+
"eval_physics_mean_token_accuracy": 0.8224700026512146,
|
| 2446 |
+
"eval_physics_num_tokens": 66193675.0,
|
| 2447 |
+
"eval_physics_runtime": 42.228,
|
| 2448 |
+
"eval_physics_samples_per_second": 11.84,
|
| 2449 |
+
"eval_physics_steps_per_second": 2.96,
|
| 2450 |
+
"step": 2000
|
| 2451 |
+
}
|
| 2452 |
+
],
|
| 2453 |
+
"logging_steps": 10,
|
| 2454 |
+
"max_steps": 10000,
|
| 2455 |
+
"num_input_tokens_seen": 0,
|
| 2456 |
+
"num_train_epochs": 16,
|
| 2457 |
+
"save_steps": 500,
|
| 2458 |
+
"stateful_callbacks": {
|
| 2459 |
+
"TrainerControl": {
|
| 2460 |
+
"args": {
|
| 2461 |
+
"should_epoch_stop": false,
|
| 2462 |
+
"should_evaluate": false,
|
| 2463 |
+
"should_log": false,
|
| 2464 |
+
"should_save": true,
|
| 2465 |
+
"should_training_stop": false
|
| 2466 |
+
},
|
| 2467 |
+
"attributes": {}
|
| 2468 |
+
}
|
| 2469 |
+
},
|
| 2470 |
+
"total_flos": 6.614072648666411e+18,
|
| 2471 |
+
"train_batch_size": 4,
|
| 2472 |
+
"trial_name": null,
|
| 2473 |
+
"trial_params": null
|
| 2474 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b77b8198261730a06a6b35e5e073ef81571e52317250e50f9755132267a61c5c
|
| 3 |
+
size 6417
|