Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- checkpoint-1500/added_tokens.json +3 -0
- checkpoint-1500/chat_template.jinja +47 -0
- checkpoint-1500/config.json +110 -0
- checkpoint-1500/generation_config.json +13 -0
- checkpoint-1500/model-00001-of-00005.safetensors +3 -0
- checkpoint-1500/model-00002-of-00005.safetensors +3 -0
- checkpoint-1500/model-00003-of-00005.safetensors +3 -0
- checkpoint-1500/model-00004-of-00005.safetensors +3 -0
- checkpoint-1500/model-00005-of-00005.safetensors +3 -0
- checkpoint-1500/model.safetensors.index.json +0 -0
- checkpoint-1500/optimizer.pt +3 -0
- checkpoint-1500/rng_state.pth +3 -0
- checkpoint-1500/scheduler.pt +3 -0
- checkpoint-1500/special_tokens_map.json +33 -0
- checkpoint-1500/tokenizer.json +3 -0
- checkpoint-1500/tokenizer.model +3 -0
- checkpoint-1500/tokenizer_config.json +0 -0
- checkpoint-1500/trainer_state.json +1864 -0
- checkpoint-1500/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 35 |
*tfevents* 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
|
|
|
|
|
|
| 35 |
*tfevents* 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
|
checkpoint-1500/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image_soft_token>": 262144
|
| 3 |
+
}
|
checkpoint-1500/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-1500/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-1500/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-1500/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-1500/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-1500/model-00003-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b15bb8eeb6f5ff843a2eb85a0872dac4e0cd6f80202972063967b1f1b2076310
|
| 3 |
+
size 4931296656
|
checkpoint-1500/model-00004-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3d0f5668738b30925773bacbf865e000f97401b13c33a6d647e4ff55a622ca4
|
| 3 |
+
size 4931296656
|
checkpoint-1500/model-00005-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c76e3cf23bfa8ca9b68e10df8b8f53976c2a804370df2c4b4ec4bf713e36b65a
|
| 3 |
+
size 4601000928
|
checkpoint-1500/model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:242dbe17a4491001404c26f669f6bf94d3d34e39b9c3ee0ac19a842159cf3810
|
| 3 |
+
size 14573116445
|
checkpoint-1500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0f035a473a5ad5cff10a645ba819c24fec2293885066a86a1d15116f370a9db
|
| 3 |
+
size 14645
|
checkpoint-1500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb8afd6405fc2f72bcdc3e89147a10597d475bf4457b8f98d9c27aa5442f3ef9
|
| 3 |
+
size 1465
|
checkpoint-1500/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-1500/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
|
| 3 |
+
size 33384568
|
checkpoint-1500/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
|
| 3 |
+
size 4689074
|
checkpoint-1500/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/trainer_state.json
ADDED
|
@@ -0,0 +1,1864 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.6,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 0.4866574574261904,
|
| 14 |
+
"epoch": 0.004,
|
| 15 |
+
"grad_norm": 616.0,
|
| 16 |
+
"learning_rate": 1.8e-07,
|
| 17 |
+
"loss": 33.2311,
|
| 18 |
+
"mean_token_accuracy": 0.6604987328872085,
|
| 19 |
+
"num_tokens": 51831.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 0.4859781404957175,
|
| 24 |
+
"epoch": 0.008,
|
| 25 |
+
"grad_norm": 676.0,
|
| 26 |
+
"learning_rate": 3.8e-07,
|
| 27 |
+
"loss": 33.1415,
|
| 28 |
+
"mean_token_accuracy": 0.6665693156421184,
|
| 29 |
+
"num_tokens": 103172.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 0.5029121663421392,
|
| 34 |
+
"epoch": 0.012,
|
| 35 |
+
"grad_norm": 824.0,
|
| 36 |
+
"learning_rate": 5.800000000000001e-07,
|
| 37 |
+
"loss": 34.7849,
|
| 38 |
+
"mean_token_accuracy": 0.6541365889832378,
|
| 39 |
+
"num_tokens": 147915.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 0.5095210419967771,
|
| 44 |
+
"epoch": 0.016,
|
| 45 |
+
"grad_norm": 588.0,
|
| 46 |
+
"learning_rate": 7.8e-07,
|
| 47 |
+
"loss": 33.2642,
|
| 48 |
+
"mean_token_accuracy": 0.6610883103683591,
|
| 49 |
+
"num_tokens": 195524.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 0.5093817379325628,
|
| 54 |
+
"epoch": 0.02,
|
| 55 |
+
"grad_norm": 544.0,
|
| 56 |
+
"learning_rate": 9.800000000000001e-07,
|
| 57 |
+
"loss": 32.1157,
|
| 58 |
+
"mean_token_accuracy": 0.6604763394221663,
|
| 59 |
+
"num_tokens": 243747.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 0.5527876092121005,
|
| 64 |
+
"epoch": 0.024,
|
| 65 |
+
"grad_norm": 628.0,
|
| 66 |
+
"learning_rate": 1.1800000000000001e-06,
|
| 67 |
+
"loss": 31.9125,
|
| 68 |
+
"mean_token_accuracy": 0.6578346157446504,
|
| 69 |
+
"num_tokens": 296999.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 0.5809885170310736,
|
| 74 |
+
"epoch": 0.028,
|
| 75 |
+
"grad_norm": 498.0,
|
| 76 |
+
"learning_rate": 1.3800000000000001e-06,
|
| 77 |
+
"loss": 32.3494,
|
| 78 |
+
"mean_token_accuracy": 0.6499254325404763,
|
| 79 |
+
"num_tokens": 342114.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 0.6243463078513741,
|
| 84 |
+
"epoch": 0.032,
|
| 85 |
+
"grad_norm": 430.0,
|
| 86 |
+
"learning_rate": 1.5800000000000001e-06,
|
| 87 |
+
"loss": 29.4405,
|
| 88 |
+
"mean_token_accuracy": 0.6568279277533293,
|
| 89 |
+
"num_tokens": 390903.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 0.7070688692852855,
|
| 94 |
+
"epoch": 0.036,
|
| 95 |
+
"grad_norm": 422.0,
|
| 96 |
+
"learning_rate": 1.7800000000000001e-06,
|
| 97 |
+
"loss": 27.9455,
|
| 98 |
+
"mean_token_accuracy": 0.6594765370711684,
|
| 99 |
+
"num_tokens": 440359.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 0.7858394492417574,
|
| 104 |
+
"epoch": 0.04,
|
| 105 |
+
"grad_norm": 334.0,
|
| 106 |
+
"learning_rate": 1.98e-06,
|
| 107 |
+
"loss": 27.3289,
|
| 108 |
+
"mean_token_accuracy": 0.6568881323561072,
|
| 109 |
+
"num_tokens": 488481.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.04,
|
| 114 |
+
"eval_biology_entropy": 0.8150426388680935,
|
| 115 |
+
"eval_biology_loss": 1.601250410079956,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.6667410479187965,
|
| 117 |
+
"eval_biology_num_tokens": 488481.0,
|
| 118 |
+
"eval_biology_runtime": 34.4305,
|
| 119 |
+
"eval_biology_samples_per_second": 14.522,
|
| 120 |
+
"eval_biology_steps_per_second": 14.522,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04,
|
| 125 |
+
"eval_chemistry_entropy": 0.42218632949888707,
|
| 126 |
+
"eval_chemistry_loss": 0.7587577700614929,
|
| 127 |
+
"eval_chemistry_mean_token_accuracy": 0.8110456941127777,
|
| 128 |
+
"eval_chemistry_num_tokens": 488481.0,
|
| 129 |
+
"eval_chemistry_runtime": 39.1702,
|
| 130 |
+
"eval_chemistry_samples_per_second": 12.765,
|
| 131 |
+
"eval_chemistry_steps_per_second": 12.765,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 0.8809404099360109,
|
| 136 |
+
"epoch": 0.044,
|
| 137 |
+
"grad_norm": 246.0,
|
| 138 |
+
"learning_rate": 2.1800000000000003e-06,
|
| 139 |
+
"loss": 25.34,
|
| 140 |
+
"mean_token_accuracy": 0.6623364068567753,
|
| 141 |
+
"num_tokens": 537471.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 0.9853538589552044,
|
| 146 |
+
"epoch": 0.048,
|
| 147 |
+
"grad_norm": 228.0,
|
| 148 |
+
"learning_rate": 2.38e-06,
|
| 149 |
+
"loss": 23.768,
|
| 150 |
+
"mean_token_accuracy": 0.6667981075122953,
|
| 151 |
+
"num_tokens": 589868.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 1.1170461051166058,
|
| 156 |
+
"epoch": 0.052,
|
| 157 |
+
"grad_norm": 171.0,
|
| 158 |
+
"learning_rate": 2.5800000000000003e-06,
|
| 159 |
+
"loss": 23.5554,
|
| 160 |
+
"mean_token_accuracy": 0.6613909218460321,
|
| 161 |
+
"num_tokens": 639512.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 1.1676542518660427,
|
| 166 |
+
"epoch": 0.056,
|
| 167 |
+
"grad_norm": 151.0,
|
| 168 |
+
"learning_rate": 2.7800000000000005e-06,
|
| 169 |
+
"loss": 22.6768,
|
| 170 |
+
"mean_token_accuracy": 0.6657767463475466,
|
| 171 |
+
"num_tokens": 690754.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 1.2725389588624239,
|
| 176 |
+
"epoch": 0.06,
|
| 177 |
+
"grad_norm": 132.0,
|
| 178 |
+
"learning_rate": 2.9800000000000003e-06,
|
| 179 |
+
"loss": 23.1722,
|
| 180 |
+
"mean_token_accuracy": 0.6627637758851052,
|
| 181 |
+
"num_tokens": 737075.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"entropy": 1.258068286627531,
|
| 186 |
+
"epoch": 0.064,
|
| 187 |
+
"grad_norm": 142.0,
|
| 188 |
+
"learning_rate": 3.1800000000000005e-06,
|
| 189 |
+
"loss": 21.9045,
|
| 190 |
+
"mean_token_accuracy": 0.6725248593837023,
|
| 191 |
+
"num_tokens": 791190.0,
|
| 192 |
+
"step": 160
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"entropy": 1.306521786376834,
|
| 196 |
+
"epoch": 0.068,
|
| 197 |
+
"grad_norm": 112.0,
|
| 198 |
+
"learning_rate": 3.3800000000000007e-06,
|
| 199 |
+
"loss": 22.1444,
|
| 200 |
+
"mean_token_accuracy": 0.6677160531282424,
|
| 201 |
+
"num_tokens": 841344.0,
|
| 202 |
+
"step": 170
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"entropy": 1.2990117866545916,
|
| 206 |
+
"epoch": 0.072,
|
| 207 |
+
"grad_norm": 141.0,
|
| 208 |
+
"learning_rate": 3.58e-06,
|
| 209 |
+
"loss": 22.1479,
|
| 210 |
+
"mean_token_accuracy": 0.6692137856036424,
|
| 211 |
+
"num_tokens": 890075.0,
|
| 212 |
+
"step": 180
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"entropy": 1.2258288636803627,
|
| 216 |
+
"epoch": 0.076,
|
| 217 |
+
"grad_norm": 99.0,
|
| 218 |
+
"learning_rate": 3.7800000000000002e-06,
|
| 219 |
+
"loss": 20.8997,
|
| 220 |
+
"mean_token_accuracy": 0.6792997987940907,
|
| 221 |
+
"num_tokens": 940599.0,
|
| 222 |
+
"step": 190
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"entropy": 1.2281328312121331,
|
| 226 |
+
"epoch": 0.08,
|
| 227 |
+
"grad_norm": 123.0,
|
| 228 |
+
"learning_rate": 3.980000000000001e-06,
|
| 229 |
+
"loss": 20.9624,
|
| 230 |
+
"mean_token_accuracy": 0.6809809722006321,
|
| 231 |
+
"num_tokens": 989860.0,
|
| 232 |
+
"step": 200
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.08,
|
| 236 |
+
"eval_biology_entropy": 1.2447693885564803,
|
| 237 |
+
"eval_biology_loss": 1.31557035446167,
|
| 238 |
+
"eval_biology_mean_token_accuracy": 0.6810677705407142,
|
| 239 |
+
"eval_biology_num_tokens": 989860.0,
|
| 240 |
+
"eval_biology_runtime": 34.2892,
|
| 241 |
+
"eval_biology_samples_per_second": 14.582,
|
| 242 |
+
"eval_biology_steps_per_second": 14.582,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.08,
|
| 247 |
+
"eval_chemistry_entropy": 0.6061330872625112,
|
| 248 |
+
"eval_chemistry_loss": 0.6521626710891724,
|
| 249 |
+
"eval_chemistry_mean_token_accuracy": 0.8165175079107284,
|
| 250 |
+
"eval_chemistry_num_tokens": 989860.0,
|
| 251 |
+
"eval_chemistry_runtime": 39.0885,
|
| 252 |
+
"eval_chemistry_samples_per_second": 12.791,
|
| 253 |
+
"eval_chemistry_steps_per_second": 12.791,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"entropy": 1.2204063970595598,
|
| 258 |
+
"epoch": 0.084,
|
| 259 |
+
"grad_norm": 99.5,
|
| 260 |
+
"learning_rate": 4.18e-06,
|
| 261 |
+
"loss": 20.4735,
|
| 262 |
+
"mean_token_accuracy": 0.688131432235241,
|
| 263 |
+
"num_tokens": 1041199.0,
|
| 264 |
+
"step": 210
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"entropy": 1.3349104586988687,
|
| 268 |
+
"epoch": 0.088,
|
| 269 |
+
"grad_norm": 111.5,
|
| 270 |
+
"learning_rate": 4.38e-06,
|
| 271 |
+
"loss": 22.5142,
|
| 272 |
+
"mean_token_accuracy": 0.6627503650262951,
|
| 273 |
+
"num_tokens": 1087099.0,
|
| 274 |
+
"step": 220
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"entropy": 1.2557024333626032,
|
| 278 |
+
"epoch": 0.092,
|
| 279 |
+
"grad_norm": 96.5,
|
| 280 |
+
"learning_rate": 4.58e-06,
|
| 281 |
+
"loss": 20.9139,
|
| 282 |
+
"mean_token_accuracy": 0.6778965264558792,
|
| 283 |
+
"num_tokens": 1135090.0,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"entropy": 1.280203009955585,
|
| 288 |
+
"epoch": 0.096,
|
| 289 |
+
"grad_norm": 105.0,
|
| 290 |
+
"learning_rate": 4.78e-06,
|
| 291 |
+
"loss": 21.3579,
|
| 292 |
+
"mean_token_accuracy": 0.6716440364718437,
|
| 293 |
+
"num_tokens": 1181777.0,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"entropy": 1.276487385854125,
|
| 298 |
+
"epoch": 0.1,
|
| 299 |
+
"grad_norm": 100.5,
|
| 300 |
+
"learning_rate": 4.980000000000001e-06,
|
| 301 |
+
"loss": 20.8968,
|
| 302 |
+
"mean_token_accuracy": 0.6772379480302334,
|
| 303 |
+
"num_tokens": 1229242.0,
|
| 304 |
+
"step": 250
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"entropy": 1.2585501682013274,
|
| 308 |
+
"epoch": 0.104,
|
| 309 |
+
"grad_norm": 85.0,
|
| 310 |
+
"learning_rate": 5.18e-06,
|
| 311 |
+
"loss": 20.7316,
|
| 312 |
+
"mean_token_accuracy": 0.679420518502593,
|
| 313 |
+
"num_tokens": 1283881.0,
|
| 314 |
+
"step": 260
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"entropy": 1.248391578346491,
|
| 318 |
+
"epoch": 0.108,
|
| 319 |
+
"grad_norm": 91.0,
|
| 320 |
+
"learning_rate": 5.380000000000001e-06,
|
| 321 |
+
"loss": 20.4527,
|
| 322 |
+
"mean_token_accuracy": 0.682051457837224,
|
| 323 |
+
"num_tokens": 1332507.0,
|
| 324 |
+
"step": 270
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"entropy": 1.2360890079289675,
|
| 328 |
+
"epoch": 0.112,
|
| 329 |
+
"grad_norm": 84.5,
|
| 330 |
+
"learning_rate": 5.580000000000001e-06,
|
| 331 |
+
"loss": 20.2916,
|
| 332 |
+
"mean_token_accuracy": 0.6859552793204784,
|
| 333 |
+
"num_tokens": 1385513.0,
|
| 334 |
+
"step": 280
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"entropy": 1.2920903146266938,
|
| 338 |
+
"epoch": 0.116,
|
| 339 |
+
"grad_norm": 90.5,
|
| 340 |
+
"learning_rate": 5.78e-06,
|
| 341 |
+
"loss": 20.9268,
|
| 342 |
+
"mean_token_accuracy": 0.673469840362668,
|
| 343 |
+
"num_tokens": 1432251.0,
|
| 344 |
+
"step": 290
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"entropy": 1.2169338904321194,
|
| 348 |
+
"epoch": 0.12,
|
| 349 |
+
"grad_norm": 87.0,
|
| 350 |
+
"learning_rate": 5.98e-06,
|
| 351 |
+
"loss": 19.6752,
|
| 352 |
+
"mean_token_accuracy": 0.6875041898339986,
|
| 353 |
+
"num_tokens": 1483667.0,
|
| 354 |
+
"step": 300
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.12,
|
| 358 |
+
"eval_biology_entropy": 1.2245318589806558,
|
| 359 |
+
"eval_biology_loss": 1.2560091018676758,
|
| 360 |
+
"eval_biology_mean_token_accuracy": 0.6886654210686683,
|
| 361 |
+
"eval_biology_num_tokens": 1483667.0,
|
| 362 |
+
"eval_biology_runtime": 34.2037,
|
| 363 |
+
"eval_biology_samples_per_second": 14.618,
|
| 364 |
+
"eval_biology_steps_per_second": 14.618,
|
| 365 |
+
"step": 300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.12,
|
| 369 |
+
"eval_chemistry_entropy": 0.6325840967297554,
|
| 370 |
+
"eval_chemistry_loss": 0.6409750580787659,
|
| 371 |
+
"eval_chemistry_mean_token_accuracy": 0.8179453414678574,
|
| 372 |
+
"eval_chemistry_num_tokens": 1483667.0,
|
| 373 |
+
"eval_chemistry_runtime": 39.0169,
|
| 374 |
+
"eval_chemistry_samples_per_second": 12.815,
|
| 375 |
+
"eval_chemistry_steps_per_second": 12.815,
|
| 376 |
+
"step": 300
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"entropy": 1.2429257260635496,
|
| 380 |
+
"epoch": 0.124,
|
| 381 |
+
"grad_norm": 79.5,
|
| 382 |
+
"learning_rate": 6.18e-06,
|
| 383 |
+
"loss": 20.6172,
|
| 384 |
+
"mean_token_accuracy": 0.6801770402118563,
|
| 385 |
+
"num_tokens": 1534286.0,
|
| 386 |
+
"step": 310
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"entropy": 1.2787830816581844,
|
| 390 |
+
"epoch": 0.128,
|
| 391 |
+
"grad_norm": 84.5,
|
| 392 |
+
"learning_rate": 6.380000000000001e-06,
|
| 393 |
+
"loss": 20.5543,
|
| 394 |
+
"mean_token_accuracy": 0.6807437939569354,
|
| 395 |
+
"num_tokens": 1582368.0,
|
| 396 |
+
"step": 320
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"entropy": 1.2588950738310813,
|
| 400 |
+
"epoch": 0.132,
|
| 401 |
+
"grad_norm": 89.0,
|
| 402 |
+
"learning_rate": 6.5800000000000005e-06,
|
| 403 |
+
"loss": 20.3303,
|
| 404 |
+
"mean_token_accuracy": 0.6814694199711084,
|
| 405 |
+
"num_tokens": 1630878.0,
|
| 406 |
+
"step": 330
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"entropy": 1.2704457983374595,
|
| 410 |
+
"epoch": 0.136,
|
| 411 |
+
"grad_norm": 83.0,
|
| 412 |
+
"learning_rate": 6.780000000000001e-06,
|
| 413 |
+
"loss": 20.6418,
|
| 414 |
+
"mean_token_accuracy": 0.6803831990808249,
|
| 415 |
+
"num_tokens": 1677060.0,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"entropy": 1.2321893341839314,
|
| 420 |
+
"epoch": 0.14,
|
| 421 |
+
"grad_norm": 87.5,
|
| 422 |
+
"learning_rate": 6.98e-06,
|
| 423 |
+
"loss": 19.8202,
|
| 424 |
+
"mean_token_accuracy": 0.6891451958566904,
|
| 425 |
+
"num_tokens": 1725186.0,
|
| 426 |
+
"step": 350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"entropy": 1.2445756699889898,
|
| 430 |
+
"epoch": 0.144,
|
| 431 |
+
"grad_norm": 92.0,
|
| 432 |
+
"learning_rate": 7.180000000000001e-06,
|
| 433 |
+
"loss": 20.2527,
|
| 434 |
+
"mean_token_accuracy": 0.6785595057532191,
|
| 435 |
+
"num_tokens": 1773764.0,
|
| 436 |
+
"step": 360
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"entropy": 1.246081338264048,
|
| 440 |
+
"epoch": 0.148,
|
| 441 |
+
"grad_norm": 82.0,
|
| 442 |
+
"learning_rate": 7.3800000000000005e-06,
|
| 443 |
+
"loss": 20.08,
|
| 444 |
+
"mean_token_accuracy": 0.6857875902205706,
|
| 445 |
+
"num_tokens": 1823706.0,
|
| 446 |
+
"step": 370
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"entropy": 1.2665387976914644,
|
| 450 |
+
"epoch": 0.152,
|
| 451 |
+
"grad_norm": 95.0,
|
| 452 |
+
"learning_rate": 7.58e-06,
|
| 453 |
+
"loss": 20.6091,
|
| 454 |
+
"mean_token_accuracy": 0.6813197182491422,
|
| 455 |
+
"num_tokens": 1873167.0,
|
| 456 |
+
"step": 380
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"entropy": 1.2548780847340821,
|
| 460 |
+
"epoch": 0.156,
|
| 461 |
+
"grad_norm": 85.0,
|
| 462 |
+
"learning_rate": 7.78e-06,
|
| 463 |
+
"loss": 20.406,
|
| 464 |
+
"mean_token_accuracy": 0.6822501845657826,
|
| 465 |
+
"num_tokens": 1921484.0,
|
| 466 |
+
"step": 390
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"entropy": 1.2407029550522566,
|
| 470 |
+
"epoch": 0.16,
|
| 471 |
+
"grad_norm": 89.0,
|
| 472 |
+
"learning_rate": 7.980000000000002e-06,
|
| 473 |
+
"loss": 19.8013,
|
| 474 |
+
"mean_token_accuracy": 0.6853796949610114,
|
| 475 |
+
"num_tokens": 1970077.0,
|
| 476 |
+
"step": 400
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.16,
|
| 480 |
+
"eval_biology_entropy": 1.214753457427025,
|
| 481 |
+
"eval_biology_loss": 1.2285655736923218,
|
| 482 |
+
"eval_biology_mean_token_accuracy": 0.6923919150233269,
|
| 483 |
+
"eval_biology_num_tokens": 1970077.0,
|
| 484 |
+
"eval_biology_runtime": 34.3823,
|
| 485 |
+
"eval_biology_samples_per_second": 14.542,
|
| 486 |
+
"eval_biology_steps_per_second": 14.542,
|
| 487 |
+
"step": 400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.16,
|
| 491 |
+
"eval_chemistry_entropy": 0.6530047204196453,
|
| 492 |
+
"eval_chemistry_loss": 0.644757866859436,
|
| 493 |
+
"eval_chemistry_mean_token_accuracy": 0.8174828389883041,
|
| 494 |
+
"eval_chemistry_num_tokens": 1970077.0,
|
| 495 |
+
"eval_chemistry_runtime": 39.0828,
|
| 496 |
+
"eval_chemistry_samples_per_second": 12.793,
|
| 497 |
+
"eval_chemistry_steps_per_second": 12.793,
|
| 498 |
+
"step": 400
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"entropy": 1.2693706121295691,
|
| 502 |
+
"epoch": 0.164,
|
| 503 |
+
"grad_norm": 75.0,
|
| 504 |
+
"learning_rate": 8.18e-06,
|
| 505 |
+
"loss": 20.5934,
|
| 506 |
+
"mean_token_accuracy": 0.6737196171656251,
|
| 507 |
+
"num_tokens": 2019053.0,
|
| 508 |
+
"step": 410
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"entropy": 1.2044812317937612,
|
| 512 |
+
"epoch": 0.168,
|
| 513 |
+
"grad_norm": 72.5,
|
| 514 |
+
"learning_rate": 8.380000000000001e-06,
|
| 515 |
+
"loss": 19.2447,
|
| 516 |
+
"mean_token_accuracy": 0.6942890204489232,
|
| 517 |
+
"num_tokens": 2066815.0,
|
| 518 |
+
"step": 420
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"entropy": 1.219953553378582,
|
| 522 |
+
"epoch": 0.172,
|
| 523 |
+
"grad_norm": 76.5,
|
| 524 |
+
"learning_rate": 8.580000000000001e-06,
|
| 525 |
+
"loss": 20.0236,
|
| 526 |
+
"mean_token_accuracy": 0.6850357944145798,
|
| 527 |
+
"num_tokens": 2116402.0,
|
| 528 |
+
"step": 430
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"entropy": 1.1823513347655534,
|
| 532 |
+
"epoch": 0.176,
|
| 533 |
+
"grad_norm": 74.0,
|
| 534 |
+
"learning_rate": 8.78e-06,
|
| 535 |
+
"loss": 19.1286,
|
| 536 |
+
"mean_token_accuracy": 0.6948613569140434,
|
| 537 |
+
"num_tokens": 2168354.0,
|
| 538 |
+
"step": 440
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"entropy": 1.2351123623549938,
|
| 542 |
+
"epoch": 0.18,
|
| 543 |
+
"grad_norm": 75.0,
|
| 544 |
+
"learning_rate": 8.98e-06,
|
| 545 |
+
"loss": 19.847,
|
| 546 |
+
"mean_token_accuracy": 0.6850900270044804,
|
| 547 |
+
"num_tokens": 2219158.0,
|
| 548 |
+
"step": 450
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"entropy": 1.2182675819844007,
|
| 552 |
+
"epoch": 0.184,
|
| 553 |
+
"grad_norm": 83.0,
|
| 554 |
+
"learning_rate": 9.180000000000002e-06,
|
| 555 |
+
"loss": 19.9054,
|
| 556 |
+
"mean_token_accuracy": 0.6874489855021239,
|
| 557 |
+
"num_tokens": 2270617.0,
|
| 558 |
+
"step": 460
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"entropy": 1.2002981722354888,
|
| 562 |
+
"epoch": 0.188,
|
| 563 |
+
"grad_norm": 74.5,
|
| 564 |
+
"learning_rate": 9.38e-06,
|
| 565 |
+
"loss": 18.9485,
|
| 566 |
+
"mean_token_accuracy": 0.6958082247525453,
|
| 567 |
+
"num_tokens": 2322034.0,
|
| 568 |
+
"step": 470
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"entropy": 1.2613376142457127,
|
| 572 |
+
"epoch": 0.192,
|
| 573 |
+
"grad_norm": 98.5,
|
| 574 |
+
"learning_rate": 9.58e-06,
|
| 575 |
+
"loss": 20.7735,
|
| 576 |
+
"mean_token_accuracy": 0.6739435123279691,
|
| 577 |
+
"num_tokens": 2365822.0,
|
| 578 |
+
"step": 480
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"entropy": 1.205817231722176,
|
| 582 |
+
"epoch": 0.196,
|
| 583 |
+
"grad_norm": 89.5,
|
| 584 |
+
"learning_rate": 9.780000000000001e-06,
|
| 585 |
+
"loss": 19.1434,
|
| 586 |
+
"mean_token_accuracy": 0.6935921791940928,
|
| 587 |
+
"num_tokens": 2413853.0,
|
| 588 |
+
"step": 490
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"entropy": 1.2515719644725323,
|
| 592 |
+
"epoch": 0.2,
|
| 593 |
+
"grad_norm": 90.5,
|
| 594 |
+
"learning_rate": 9.980000000000001e-06,
|
| 595 |
+
"loss": 20.4938,
|
| 596 |
+
"mean_token_accuracy": 0.6777064654976129,
|
| 597 |
+
"num_tokens": 2459876.0,
|
| 598 |
+
"step": 500
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.2,
|
| 602 |
+
"eval_biology_entropy": 1.1872086516022682,
|
| 603 |
+
"eval_biology_loss": 1.2085970640182495,
|
| 604 |
+
"eval_biology_mean_token_accuracy": 0.6949169620871544,
|
| 605 |
+
"eval_biology_num_tokens": 2459876.0,
|
| 606 |
+
"eval_biology_runtime": 34.4,
|
| 607 |
+
"eval_biology_samples_per_second": 14.535,
|
| 608 |
+
"eval_biology_steps_per_second": 14.535,
|
| 609 |
+
"step": 500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.2,
|
| 613 |
+
"eval_chemistry_entropy": 0.6488757311999798,
|
| 614 |
+
"eval_chemistry_loss": 0.6450773477554321,
|
| 615 |
+
"eval_chemistry_mean_token_accuracy": 0.8178821201324463,
|
| 616 |
+
"eval_chemistry_num_tokens": 2459876.0,
|
| 617 |
+
"eval_chemistry_runtime": 39.2328,
|
| 618 |
+
"eval_chemistry_samples_per_second": 12.744,
|
| 619 |
+
"eval_chemistry_steps_per_second": 12.744,
|
| 620 |
+
"step": 500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"entropy": 1.2209952611476182,
|
| 624 |
+
"epoch": 0.204,
|
| 625 |
+
"grad_norm": 73.0,
|
| 626 |
+
"learning_rate": 1.018e-05,
|
| 627 |
+
"loss": 19.6626,
|
| 628 |
+
"mean_token_accuracy": 0.6881283605471253,
|
| 629 |
+
"num_tokens": 2509187.0,
|
| 630 |
+
"step": 510
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"entropy": 1.2497611749917268,
|
| 634 |
+
"epoch": 0.208,
|
| 635 |
+
"grad_norm": 85.5,
|
| 636 |
+
"learning_rate": 1.038e-05,
|
| 637 |
+
"loss": 19.9518,
|
| 638 |
+
"mean_token_accuracy": 0.6810324152931571,
|
| 639 |
+
"num_tokens": 2558762.0,
|
| 640 |
+
"step": 520
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"entropy": 1.1757056951522826,
|
| 644 |
+
"epoch": 0.212,
|
| 645 |
+
"grad_norm": 78.0,
|
| 646 |
+
"learning_rate": 1.0580000000000002e-05,
|
| 647 |
+
"loss": 19.3754,
|
| 648 |
+
"mean_token_accuracy": 0.6935973027721047,
|
| 649 |
+
"num_tokens": 2608616.0,
|
| 650 |
+
"step": 530
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"entropy": 1.2561812430620194,
|
| 654 |
+
"epoch": 0.216,
|
| 655 |
+
"grad_norm": 83.0,
|
| 656 |
+
"learning_rate": 1.0780000000000002e-05,
|
| 657 |
+
"loss": 20.0513,
|
| 658 |
+
"mean_token_accuracy": 0.6874226013198494,
|
| 659 |
+
"num_tokens": 2654953.0,
|
| 660 |
+
"step": 540
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"entropy": 1.2020427081733942,
|
| 664 |
+
"epoch": 0.22,
|
| 665 |
+
"grad_norm": 78.5,
|
| 666 |
+
"learning_rate": 1.0980000000000002e-05,
|
| 667 |
+
"loss": 19.5412,
|
| 668 |
+
"mean_token_accuracy": 0.688438569009304,
|
| 669 |
+
"num_tokens": 2704686.0,
|
| 670 |
+
"step": 550
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"entropy": 1.205025901272893,
|
| 674 |
+
"epoch": 0.224,
|
| 675 |
+
"grad_norm": 86.5,
|
| 676 |
+
"learning_rate": 1.1180000000000001e-05,
|
| 677 |
+
"loss": 19.7851,
|
| 678 |
+
"mean_token_accuracy": 0.6885192114859819,
|
| 679 |
+
"num_tokens": 2755347.0,
|
| 680 |
+
"step": 560
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"entropy": 1.2252822391688825,
|
| 684 |
+
"epoch": 0.228,
|
| 685 |
+
"grad_norm": 74.5,
|
| 686 |
+
"learning_rate": 1.138e-05,
|
| 687 |
+
"loss": 19.3164,
|
| 688 |
+
"mean_token_accuracy": 0.6909766353666782,
|
| 689 |
+
"num_tokens": 2804340.0,
|
| 690 |
+
"step": 570
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"entropy": 1.1826427452266217,
|
| 694 |
+
"epoch": 0.232,
|
| 695 |
+
"grad_norm": 80.0,
|
| 696 |
+
"learning_rate": 1.1580000000000001e-05,
|
| 697 |
+
"loss": 19.3402,
|
| 698 |
+
"mean_token_accuracy": 0.6917916279286146,
|
| 699 |
+
"num_tokens": 2852298.0,
|
| 700 |
+
"step": 580
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"entropy": 1.1695974227041006,
|
| 704 |
+
"epoch": 0.236,
|
| 705 |
+
"grad_norm": 89.0,
|
| 706 |
+
"learning_rate": 1.178e-05,
|
| 707 |
+
"loss": 19.0702,
|
| 708 |
+
"mean_token_accuracy": 0.6989134319126606,
|
| 709 |
+
"num_tokens": 2901227.0,
|
| 710 |
+
"step": 590
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"entropy": 1.2379697296768426,
|
| 714 |
+
"epoch": 0.24,
|
| 715 |
+
"grad_norm": 92.5,
|
| 716 |
+
"learning_rate": 1.198e-05,
|
| 717 |
+
"loss": 19.6279,
|
| 718 |
+
"mean_token_accuracy": 0.6916081016883254,
|
| 719 |
+
"num_tokens": 2947346.0,
|
| 720 |
+
"step": 600
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.24,
|
| 724 |
+
"eval_biology_entropy": 1.1614825129508972,
|
| 725 |
+
"eval_biology_loss": 1.197817087173462,
|
| 726 |
+
"eval_biology_mean_token_accuracy": 0.6965834299921989,
|
| 727 |
+
"eval_biology_num_tokens": 2947346.0,
|
| 728 |
+
"eval_biology_runtime": 34.3297,
|
| 729 |
+
"eval_biology_samples_per_second": 14.565,
|
| 730 |
+
"eval_biology_steps_per_second": 14.565,
|
| 731 |
+
"step": 600
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.24,
|
| 735 |
+
"eval_chemistry_entropy": 0.6421602384448052,
|
| 736 |
+
"eval_chemistry_loss": 0.6457399725914001,
|
| 737 |
+
"eval_chemistry_mean_token_accuracy": 0.8177337770462036,
|
| 738 |
+
"eval_chemistry_num_tokens": 2947346.0,
|
| 739 |
+
"eval_chemistry_runtime": 39.0493,
|
| 740 |
+
"eval_chemistry_samples_per_second": 12.804,
|
| 741 |
+
"eval_chemistry_steps_per_second": 12.804,
|
| 742 |
+
"step": 600
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"entropy": 1.2068689458072186,
|
| 746 |
+
"epoch": 0.244,
|
| 747 |
+
"grad_norm": 83.0,
|
| 748 |
+
"learning_rate": 1.218e-05,
|
| 749 |
+
"loss": 19.9725,
|
| 750 |
+
"mean_token_accuracy": 0.6830457381904125,
|
| 751 |
+
"num_tokens": 2996175.0,
|
| 752 |
+
"step": 610
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"entropy": 1.2800738830119371,
|
| 756 |
+
"epoch": 0.248,
|
| 757 |
+
"grad_norm": 77.5,
|
| 758 |
+
"learning_rate": 1.2380000000000002e-05,
|
| 759 |
+
"loss": 20.1671,
|
| 760 |
+
"mean_token_accuracy": 0.6834735475480557,
|
| 761 |
+
"num_tokens": 3043128.0,
|
| 762 |
+
"step": 620
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"entropy": 1.190559478290379,
|
| 766 |
+
"epoch": 0.252,
|
| 767 |
+
"grad_norm": 79.5,
|
| 768 |
+
"learning_rate": 1.2580000000000002e-05,
|
| 769 |
+
"loss": 19.3606,
|
| 770 |
+
"mean_token_accuracy": 0.6899204226210713,
|
| 771 |
+
"num_tokens": 3090827.0,
|
| 772 |
+
"step": 630
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"entropy": 1.1722685299813747,
|
| 776 |
+
"epoch": 0.256,
|
| 777 |
+
"grad_norm": 81.5,
|
| 778 |
+
"learning_rate": 1.2780000000000001e-05,
|
| 779 |
+
"loss": 18.8652,
|
| 780 |
+
"mean_token_accuracy": 0.6963184822350741,
|
| 781 |
+
"num_tokens": 3139957.0,
|
| 782 |
+
"step": 640
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"entropy": 1.2144603542983532,
|
| 786 |
+
"epoch": 0.26,
|
| 787 |
+
"grad_norm": 82.0,
|
| 788 |
+
"learning_rate": 1.2980000000000001e-05,
|
| 789 |
+
"loss": 19.7174,
|
| 790 |
+
"mean_token_accuracy": 0.685292916931212,
|
| 791 |
+
"num_tokens": 3183188.0,
|
| 792 |
+
"step": 650
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"entropy": 1.2338291574269533,
|
| 796 |
+
"epoch": 0.264,
|
| 797 |
+
"grad_norm": 80.0,
|
| 798 |
+
"learning_rate": 1.3180000000000001e-05,
|
| 799 |
+
"loss": 19.7757,
|
| 800 |
+
"mean_token_accuracy": 0.6882328724488616,
|
| 801 |
+
"num_tokens": 3232159.0,
|
| 802 |
+
"step": 660
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"entropy": 1.1896044220775366,
|
| 806 |
+
"epoch": 0.268,
|
| 807 |
+
"grad_norm": 79.5,
|
| 808 |
+
"learning_rate": 1.3380000000000002e-05,
|
| 809 |
+
"loss": 19.1149,
|
| 810 |
+
"mean_token_accuracy": 0.6937362670898437,
|
| 811 |
+
"num_tokens": 3283908.0,
|
| 812 |
+
"step": 670
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"entropy": 1.1706442803144455,
|
| 816 |
+
"epoch": 0.272,
|
| 817 |
+
"grad_norm": 98.0,
|
| 818 |
+
"learning_rate": 1.3580000000000002e-05,
|
| 819 |
+
"loss": 19.1832,
|
| 820 |
+
"mean_token_accuracy": 0.6947279252111912,
|
| 821 |
+
"num_tokens": 3335951.0,
|
| 822 |
+
"step": 680
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"entropy": 1.2103422859683632,
|
| 826 |
+
"epoch": 0.276,
|
| 827 |
+
"grad_norm": 88.0,
|
| 828 |
+
"learning_rate": 1.378e-05,
|
| 829 |
+
"loss": 19.5378,
|
| 830 |
+
"mean_token_accuracy": 0.6888058576732874,
|
| 831 |
+
"num_tokens": 3387755.0,
|
| 832 |
+
"step": 690
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"entropy": 1.1900839667767287,
|
| 836 |
+
"epoch": 0.28,
|
| 837 |
+
"grad_norm": 90.0,
|
| 838 |
+
"learning_rate": 1.398e-05,
|
| 839 |
+
"loss": 19.3154,
|
| 840 |
+
"mean_token_accuracy": 0.6907492563128471,
|
| 841 |
+
"num_tokens": 3437213.0,
|
| 842 |
+
"step": 700
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.28,
|
| 846 |
+
"eval_biology_entropy": 1.1723266056776047,
|
| 847 |
+
"eval_biology_loss": 1.1935571432113647,
|
| 848 |
+
"eval_biology_mean_token_accuracy": 0.6974302505254746,
|
| 849 |
+
"eval_biology_num_tokens": 3437213.0,
|
| 850 |
+
"eval_biology_runtime": 34.4282,
|
| 851 |
+
"eval_biology_samples_per_second": 14.523,
|
| 852 |
+
"eval_biology_steps_per_second": 14.523,
|
| 853 |
+
"step": 700
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 0.28,
|
| 857 |
+
"eval_chemistry_entropy": 0.6567539082467556,
|
| 858 |
+
"eval_chemistry_loss": 0.6476752758026123,
|
| 859 |
+
"eval_chemistry_mean_token_accuracy": 0.8169361530542374,
|
| 860 |
+
"eval_chemistry_num_tokens": 3437213.0,
|
| 861 |
+
"eval_chemistry_runtime": 39.1281,
|
| 862 |
+
"eval_chemistry_samples_per_second": 12.779,
|
| 863 |
+
"eval_chemistry_steps_per_second": 12.779,
|
| 864 |
+
"step": 700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"entropy": 1.1601122960448265,
|
| 868 |
+
"epoch": 0.284,
|
| 869 |
+
"grad_norm": 70.0,
|
| 870 |
+
"learning_rate": 1.418e-05,
|
| 871 |
+
"loss": 18.4856,
|
| 872 |
+
"mean_token_accuracy": 0.7002454694360495,
|
| 873 |
+
"num_tokens": 3489088.0,
|
| 874 |
+
"step": 710
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"entropy": 1.1836338724941016,
|
| 878 |
+
"epoch": 0.288,
|
| 879 |
+
"grad_norm": 74.0,
|
| 880 |
+
"learning_rate": 1.4380000000000001e-05,
|
| 881 |
+
"loss": 19.4318,
|
| 882 |
+
"mean_token_accuracy": 0.6893038578331471,
|
| 883 |
+
"num_tokens": 3539731.0,
|
| 884 |
+
"step": 720
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"entropy": 1.215116088837385,
|
| 888 |
+
"epoch": 0.292,
|
| 889 |
+
"grad_norm": 76.5,
|
| 890 |
+
"learning_rate": 1.4580000000000001e-05,
|
| 891 |
+
"loss": 19.2441,
|
| 892 |
+
"mean_token_accuracy": 0.6934854052960873,
|
| 893 |
+
"num_tokens": 3586433.0,
|
| 894 |
+
"step": 730
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"entropy": 1.2150515001267195,
|
| 898 |
+
"epoch": 0.296,
|
| 899 |
+
"grad_norm": 87.5,
|
| 900 |
+
"learning_rate": 1.478e-05,
|
| 901 |
+
"loss": 20.0166,
|
| 902 |
+
"mean_token_accuracy": 0.6854536816477775,
|
| 903 |
+
"num_tokens": 3633448.0,
|
| 904 |
+
"step": 740
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"entropy": 1.1540303017944098,
|
| 908 |
+
"epoch": 0.3,
|
| 909 |
+
"grad_norm": 77.5,
|
| 910 |
+
"learning_rate": 1.498e-05,
|
| 911 |
+
"loss": 18.4485,
|
| 912 |
+
"mean_token_accuracy": 0.705385484918952,
|
| 913 |
+
"num_tokens": 3685080.0,
|
| 914 |
+
"step": 750
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"entropy": 1.1504614014178514,
|
| 918 |
+
"epoch": 0.304,
|
| 919 |
+
"grad_norm": 79.0,
|
| 920 |
+
"learning_rate": 1.5180000000000002e-05,
|
| 921 |
+
"loss": 18.9317,
|
| 922 |
+
"mean_token_accuracy": 0.6943564370274544,
|
| 923 |
+
"num_tokens": 3733488.0,
|
| 924 |
+
"step": 760
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"entropy": 1.227951274625957,
|
| 928 |
+
"epoch": 0.308,
|
| 929 |
+
"grad_norm": 81.0,
|
| 930 |
+
"learning_rate": 1.5380000000000002e-05,
|
| 931 |
+
"loss": 19.6245,
|
| 932 |
+
"mean_token_accuracy": 0.69070417098701,
|
| 933 |
+
"num_tokens": 3779708.0,
|
| 934 |
+
"step": 770
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"entropy": 1.1914822444319726,
|
| 938 |
+
"epoch": 0.312,
|
| 939 |
+
"grad_norm": 91.5,
|
| 940 |
+
"learning_rate": 1.5580000000000003e-05,
|
| 941 |
+
"loss": 19.3053,
|
| 942 |
+
"mean_token_accuracy": 0.691599304229021,
|
| 943 |
+
"num_tokens": 3825212.0,
|
| 944 |
+
"step": 780
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"entropy": 1.2374459221959113,
|
| 948 |
+
"epoch": 0.316,
|
| 949 |
+
"grad_norm": 90.5,
|
| 950 |
+
"learning_rate": 1.578e-05,
|
| 951 |
+
"loss": 20.1615,
|
| 952 |
+
"mean_token_accuracy": 0.6793428828939796,
|
| 953 |
+
"num_tokens": 3870912.0,
|
| 954 |
+
"step": 790
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"entropy": 1.204637792892754,
|
| 958 |
+
"epoch": 0.32,
|
| 959 |
+
"grad_norm": 87.0,
|
| 960 |
+
"learning_rate": 1.5980000000000003e-05,
|
| 961 |
+
"loss": 19.1533,
|
| 962 |
+
"mean_token_accuracy": 0.6921680744737386,
|
| 963 |
+
"num_tokens": 3920545.0,
|
| 964 |
+
"step": 800
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 0.32,
|
| 968 |
+
"eval_biology_entropy": 1.1595962298512459,
|
| 969 |
+
"eval_biology_loss": 1.1908730268478394,
|
| 970 |
+
"eval_biology_mean_token_accuracy": 0.6982393438220024,
|
| 971 |
+
"eval_biology_num_tokens": 3920545.0,
|
| 972 |
+
"eval_biology_runtime": 34.3027,
|
| 973 |
+
"eval_biology_samples_per_second": 14.576,
|
| 974 |
+
"eval_biology_steps_per_second": 14.576,
|
| 975 |
+
"step": 800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.32,
|
| 979 |
+
"eval_chemistry_entropy": 0.6581202555298805,
|
| 980 |
+
"eval_chemistry_loss": 0.6543570756912231,
|
| 981 |
+
"eval_chemistry_mean_token_accuracy": 0.8150564895868302,
|
| 982 |
+
"eval_chemistry_num_tokens": 3920545.0,
|
| 983 |
+
"eval_chemistry_runtime": 39.1221,
|
| 984 |
+
"eval_chemistry_samples_per_second": 12.781,
|
| 985 |
+
"eval_chemistry_steps_per_second": 12.781,
|
| 986 |
+
"step": 800
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"entropy": 1.1894057728350163,
|
| 990 |
+
"epoch": 0.324,
|
| 991 |
+
"grad_norm": 82.0,
|
| 992 |
+
"learning_rate": 1.618e-05,
|
| 993 |
+
"loss": 19.3278,
|
| 994 |
+
"mean_token_accuracy": 0.6910803083330392,
|
| 995 |
+
"num_tokens": 3967499.0,
|
| 996 |
+
"step": 810
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"entropy": 1.1689968170598148,
|
| 1000 |
+
"epoch": 0.328,
|
| 1001 |
+
"grad_norm": 75.5,
|
| 1002 |
+
"learning_rate": 1.638e-05,
|
| 1003 |
+
"loss": 18.6792,
|
| 1004 |
+
"mean_token_accuracy": 0.6996502809226512,
|
| 1005 |
+
"num_tokens": 4016673.0,
|
| 1006 |
+
"step": 820
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"entropy": 1.1961608117446303,
|
| 1010 |
+
"epoch": 0.332,
|
| 1011 |
+
"grad_norm": 88.5,
|
| 1012 |
+
"learning_rate": 1.658e-05,
|
| 1013 |
+
"loss": 19.6175,
|
| 1014 |
+
"mean_token_accuracy": 0.6912782397121191,
|
| 1015 |
+
"num_tokens": 4063859.0,
|
| 1016 |
+
"step": 830
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"entropy": 1.2022534670308231,
|
| 1020 |
+
"epoch": 0.336,
|
| 1021 |
+
"grad_norm": 75.5,
|
| 1022 |
+
"learning_rate": 1.6780000000000002e-05,
|
| 1023 |
+
"loss": 19.3189,
|
| 1024 |
+
"mean_token_accuracy": 0.6922560017555952,
|
| 1025 |
+
"num_tokens": 4114077.0,
|
| 1026 |
+
"step": 840
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"entropy": 1.2139138951897621,
|
| 1030 |
+
"epoch": 0.34,
|
| 1031 |
+
"grad_norm": 80.0,
|
| 1032 |
+
"learning_rate": 1.698e-05,
|
| 1033 |
+
"loss": 19.6407,
|
| 1034 |
+
"mean_token_accuracy": 0.6891672356054187,
|
| 1035 |
+
"num_tokens": 4160168.0,
|
| 1036 |
+
"step": 850
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"entropy": 1.1922080295160413,
|
| 1040 |
+
"epoch": 0.344,
|
| 1041 |
+
"grad_norm": 76.5,
|
| 1042 |
+
"learning_rate": 1.718e-05,
|
| 1043 |
+
"loss": 19.3527,
|
| 1044 |
+
"mean_token_accuracy": 0.6934164728969335,
|
| 1045 |
+
"num_tokens": 4210403.0,
|
| 1046 |
+
"step": 860
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"entropy": 1.177280587144196,
|
| 1050 |
+
"epoch": 0.348,
|
| 1051 |
+
"grad_norm": 76.0,
|
| 1052 |
+
"learning_rate": 1.7380000000000003e-05,
|
| 1053 |
+
"loss": 18.7411,
|
| 1054 |
+
"mean_token_accuracy": 0.6976979669183493,
|
| 1055 |
+
"num_tokens": 4256944.0,
|
| 1056 |
+
"step": 870
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"entropy": 1.2034492146223783,
|
| 1060 |
+
"epoch": 0.352,
|
| 1061 |
+
"grad_norm": 77.5,
|
| 1062 |
+
"learning_rate": 1.758e-05,
|
| 1063 |
+
"loss": 19.6262,
|
| 1064 |
+
"mean_token_accuracy": 0.6897061046212911,
|
| 1065 |
+
"num_tokens": 4306949.0,
|
| 1066 |
+
"step": 880
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"entropy": 1.2024015821516514,
|
| 1070 |
+
"epoch": 0.356,
|
| 1071 |
+
"grad_norm": 70.5,
|
| 1072 |
+
"learning_rate": 1.7780000000000003e-05,
|
| 1073 |
+
"loss": 19.4871,
|
| 1074 |
+
"mean_token_accuracy": 0.6887454513460398,
|
| 1075 |
+
"num_tokens": 4355070.0,
|
| 1076 |
+
"step": 890
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"entropy": 1.1949962507933378,
|
| 1080 |
+
"epoch": 0.36,
|
| 1081 |
+
"grad_norm": 65.0,
|
| 1082 |
+
"learning_rate": 1.798e-05,
|
| 1083 |
+
"loss": 19.252,
|
| 1084 |
+
"mean_token_accuracy": 0.6892035495489836,
|
| 1085 |
+
"num_tokens": 4406104.0,
|
| 1086 |
+
"step": 900
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.36,
|
| 1090 |
+
"eval_biology_entropy": 1.149014304101467,
|
| 1091 |
+
"eval_biology_loss": 1.191479206085205,
|
| 1092 |
+
"eval_biology_mean_token_accuracy": 0.6979464264512062,
|
| 1093 |
+
"eval_biology_num_tokens": 4406104.0,
|
| 1094 |
+
"eval_biology_runtime": 34.5339,
|
| 1095 |
+
"eval_biology_samples_per_second": 14.479,
|
| 1096 |
+
"eval_biology_steps_per_second": 14.479,
|
| 1097 |
+
"step": 900
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.36,
|
| 1101 |
+
"eval_chemistry_entropy": 0.6591544916331769,
|
| 1102 |
+
"eval_chemistry_loss": 0.6548386812210083,
|
| 1103 |
+
"eval_chemistry_mean_token_accuracy": 0.8150800436735153,
|
| 1104 |
+
"eval_chemistry_num_tokens": 4406104.0,
|
| 1105 |
+
"eval_chemistry_runtime": 39.1161,
|
| 1106 |
+
"eval_chemistry_samples_per_second": 12.782,
|
| 1107 |
+
"eval_chemistry_steps_per_second": 12.782,
|
| 1108 |
+
"step": 900
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"entropy": 1.210719463415444,
|
| 1112 |
+
"epoch": 0.364,
|
| 1113 |
+
"grad_norm": 77.5,
|
| 1114 |
+
"learning_rate": 1.8180000000000002e-05,
|
| 1115 |
+
"loss": 19.6229,
|
| 1116 |
+
"mean_token_accuracy": 0.6920540288090706,
|
| 1117 |
+
"num_tokens": 4455015.0,
|
| 1118 |
+
"step": 910
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"entropy": 1.1651359049603343,
|
| 1122 |
+
"epoch": 0.368,
|
| 1123 |
+
"grad_norm": 85.0,
|
| 1124 |
+
"learning_rate": 1.8380000000000004e-05,
|
| 1125 |
+
"loss": 18.965,
|
| 1126 |
+
"mean_token_accuracy": 0.6945374647155405,
|
| 1127 |
+
"num_tokens": 4504001.0,
|
| 1128 |
+
"step": 920
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"entropy": 1.2309420805424451,
|
| 1132 |
+
"epoch": 0.372,
|
| 1133 |
+
"grad_norm": 79.0,
|
| 1134 |
+
"learning_rate": 1.858e-05,
|
| 1135 |
+
"loss": 19.4645,
|
| 1136 |
+
"mean_token_accuracy": 0.6926089435815811,
|
| 1137 |
+
"num_tokens": 4550269.0,
|
| 1138 |
+
"step": 930
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"entropy": 1.1543804241344333,
|
| 1142 |
+
"epoch": 0.376,
|
| 1143 |
+
"grad_norm": 83.5,
|
| 1144 |
+
"learning_rate": 1.878e-05,
|
| 1145 |
+
"loss": 18.9949,
|
| 1146 |
+
"mean_token_accuracy": 0.695966462418437,
|
| 1147 |
+
"num_tokens": 4597559.0,
|
| 1148 |
+
"step": 940
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"entropy": 1.1860109277069568,
|
| 1152 |
+
"epoch": 0.38,
|
| 1153 |
+
"grad_norm": 88.0,
|
| 1154 |
+
"learning_rate": 1.898e-05,
|
| 1155 |
+
"loss": 19.0122,
|
| 1156 |
+
"mean_token_accuracy": 0.6945275101810694,
|
| 1157 |
+
"num_tokens": 4646903.0,
|
| 1158 |
+
"step": 950
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"entropy": 1.1854952551424502,
|
| 1162 |
+
"epoch": 0.384,
|
| 1163 |
+
"grad_norm": 72.0,
|
| 1164 |
+
"learning_rate": 1.918e-05,
|
| 1165 |
+
"loss": 19.3612,
|
| 1166 |
+
"mean_token_accuracy": 0.6901288989931345,
|
| 1167 |
+
"num_tokens": 4693812.0,
|
| 1168 |
+
"step": 960
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"entropy": 1.2153874900192023,
|
| 1172 |
+
"epoch": 0.388,
|
| 1173 |
+
"grad_norm": 76.0,
|
| 1174 |
+
"learning_rate": 1.938e-05,
|
| 1175 |
+
"loss": 19.4325,
|
| 1176 |
+
"mean_token_accuracy": 0.6911413293331862,
|
| 1177 |
+
"num_tokens": 4740309.0,
|
| 1178 |
+
"step": 970
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"entropy": 1.212491791881621,
|
| 1182 |
+
"epoch": 0.392,
|
| 1183 |
+
"grad_norm": 74.0,
|
| 1184 |
+
"learning_rate": 1.9580000000000002e-05,
|
| 1185 |
+
"loss": 19.43,
|
| 1186 |
+
"mean_token_accuracy": 0.6912798311561346,
|
| 1187 |
+
"num_tokens": 4790965.0,
|
| 1188 |
+
"step": 980
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"entropy": 1.1938600786030293,
|
| 1192 |
+
"epoch": 0.396,
|
| 1193 |
+
"grad_norm": 77.0,
|
| 1194 |
+
"learning_rate": 1.978e-05,
|
| 1195 |
+
"loss": 19.4531,
|
| 1196 |
+
"mean_token_accuracy": 0.6905827283859253,
|
| 1197 |
+
"num_tokens": 4839186.0,
|
| 1198 |
+
"step": 990
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"entropy": 1.2035906264558434,
|
| 1202 |
+
"epoch": 0.4,
|
| 1203 |
+
"grad_norm": 82.5,
|
| 1204 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 1205 |
+
"loss": 19.4249,
|
| 1206 |
+
"mean_token_accuracy": 0.6903344210237264,
|
| 1207 |
+
"num_tokens": 4887094.0,
|
| 1208 |
+
"step": 1000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.4,
|
| 1212 |
+
"eval_biology_entropy": 1.1641007453799248,
|
| 1213 |
+
"eval_biology_loss": 1.1950935125350952,
|
| 1214 |
+
"eval_biology_mean_token_accuracy": 0.6969818140268326,
|
| 1215 |
+
"eval_biology_num_tokens": 4887094.0,
|
| 1216 |
+
"eval_biology_runtime": 34.3544,
|
| 1217 |
+
"eval_biology_samples_per_second": 14.554,
|
| 1218 |
+
"eval_biology_steps_per_second": 14.554,
|
| 1219 |
+
"step": 1000
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.4,
|
| 1223 |
+
"eval_chemistry_entropy": 0.6638510838449001,
|
| 1224 |
+
"eval_chemistry_loss": 0.662416934967041,
|
| 1225 |
+
"eval_chemistry_mean_token_accuracy": 0.8127187831401825,
|
| 1226 |
+
"eval_chemistry_num_tokens": 4887094.0,
|
| 1227 |
+
"eval_chemistry_runtime": 38.9213,
|
| 1228 |
+
"eval_chemistry_samples_per_second": 12.846,
|
| 1229 |
+
"eval_chemistry_steps_per_second": 12.846,
|
| 1230 |
+
"step": 1000
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"entropy": 1.2332831894978882,
|
| 1234 |
+
"epoch": 0.404,
|
| 1235 |
+
"grad_norm": 88.5,
|
| 1236 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 1237 |
+
"loss": 19.9936,
|
| 1238 |
+
"mean_token_accuracy": 0.6858718886971473,
|
| 1239 |
+
"num_tokens": 4932228.0,
|
| 1240 |
+
"step": 1010
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"entropy": 1.1915043991059064,
|
| 1244 |
+
"epoch": 0.408,
|
| 1245 |
+
"grad_norm": 64.5,
|
| 1246 |
+
"learning_rate": 1.995777777777778e-05,
|
| 1247 |
+
"loss": 19.108,
|
| 1248 |
+
"mean_token_accuracy": 0.6954168599098921,
|
| 1249 |
+
"num_tokens": 4985394.0,
|
| 1250 |
+
"step": 1020
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"entropy": 1.2241652118042112,
|
| 1254 |
+
"epoch": 0.412,
|
| 1255 |
+
"grad_norm": 65.0,
|
| 1256 |
+
"learning_rate": 1.9935555555555557e-05,
|
| 1257 |
+
"loss": 19.765,
|
| 1258 |
+
"mean_token_accuracy": 0.6827985186129808,
|
| 1259 |
+
"num_tokens": 5037274.0,
|
| 1260 |
+
"step": 1030
|
| 1261 |
+
},
|
| 1262 |
+
{
|
| 1263 |
+
"entropy": 1.207521514967084,
|
| 1264 |
+
"epoch": 0.416,
|
| 1265 |
+
"grad_norm": 65.0,
|
| 1266 |
+
"learning_rate": 1.9913333333333335e-05,
|
| 1267 |
+
"loss": 19.6241,
|
| 1268 |
+
"mean_token_accuracy": 0.6886004239320755,
|
| 1269 |
+
"num_tokens": 5085369.0,
|
| 1270 |
+
"step": 1040
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"entropy": 1.2827957067638636,
|
| 1274 |
+
"epoch": 0.42,
|
| 1275 |
+
"grad_norm": 90.0,
|
| 1276 |
+
"learning_rate": 1.9891111111111112e-05,
|
| 1277 |
+
"loss": 20.5645,
|
| 1278 |
+
"mean_token_accuracy": 0.6782789714634418,
|
| 1279 |
+
"num_tokens": 5129587.0,
|
| 1280 |
+
"step": 1050
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"entropy": 1.2159279894083739,
|
| 1284 |
+
"epoch": 0.424,
|
| 1285 |
+
"grad_norm": 70.5,
|
| 1286 |
+
"learning_rate": 1.986888888888889e-05,
|
| 1287 |
+
"loss": 19.7117,
|
| 1288 |
+
"mean_token_accuracy": 0.6862207680940628,
|
| 1289 |
+
"num_tokens": 5176083.0,
|
| 1290 |
+
"step": 1060
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"entropy": 1.1846632236614822,
|
| 1294 |
+
"epoch": 0.428,
|
| 1295 |
+
"grad_norm": 83.5,
|
| 1296 |
+
"learning_rate": 1.9846666666666668e-05,
|
| 1297 |
+
"loss": 19.1506,
|
| 1298 |
+
"mean_token_accuracy": 0.6937350874766708,
|
| 1299 |
+
"num_tokens": 5225296.0,
|
| 1300 |
+
"step": 1070
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"entropy": 1.2120982160791756,
|
| 1304 |
+
"epoch": 0.432,
|
| 1305 |
+
"grad_norm": 81.0,
|
| 1306 |
+
"learning_rate": 1.9824444444444445e-05,
|
| 1307 |
+
"loss": 19.3732,
|
| 1308 |
+
"mean_token_accuracy": 0.6929085306823254,
|
| 1309 |
+
"num_tokens": 5271275.0,
|
| 1310 |
+
"step": 1080
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"entropy": 1.216021078079939,
|
| 1314 |
+
"epoch": 0.436,
|
| 1315 |
+
"grad_norm": 74.0,
|
| 1316 |
+
"learning_rate": 1.9802222222222226e-05,
|
| 1317 |
+
"loss": 19.6442,
|
| 1318 |
+
"mean_token_accuracy": 0.6861835535615682,
|
| 1319 |
+
"num_tokens": 5318781.0,
|
| 1320 |
+
"step": 1090
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"entropy": 1.2075613098219038,
|
| 1324 |
+
"epoch": 0.44,
|
| 1325 |
+
"grad_norm": 74.5,
|
| 1326 |
+
"learning_rate": 1.978e-05,
|
| 1327 |
+
"loss": 19.5288,
|
| 1328 |
+
"mean_token_accuracy": 0.6872704658657313,
|
| 1329 |
+
"num_tokens": 5362624.0,
|
| 1330 |
+
"step": 1100
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.44,
|
| 1334 |
+
"eval_biology_entropy": 1.2076813784837723,
|
| 1335 |
+
"eval_biology_loss": 1.1966326236724854,
|
| 1336 |
+
"eval_biology_mean_token_accuracy": 0.6968241932392121,
|
| 1337 |
+
"eval_biology_num_tokens": 5362624.0,
|
| 1338 |
+
"eval_biology_runtime": 34.4319,
|
| 1339 |
+
"eval_biology_samples_per_second": 14.521,
|
| 1340 |
+
"eval_biology_steps_per_second": 14.521,
|
| 1341 |
+
"step": 1100
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 0.44,
|
| 1345 |
+
"eval_chemistry_entropy": 0.6861711817383767,
|
| 1346 |
+
"eval_chemistry_loss": 0.6706647872924805,
|
| 1347 |
+
"eval_chemistry_mean_token_accuracy": 0.8111056753396988,
|
| 1348 |
+
"eval_chemistry_num_tokens": 5362624.0,
|
| 1349 |
+
"eval_chemistry_runtime": 39.1421,
|
| 1350 |
+
"eval_chemistry_samples_per_second": 12.774,
|
| 1351 |
+
"eval_chemistry_steps_per_second": 12.774,
|
| 1352 |
+
"step": 1100
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"entropy": 1.2112580770626664,
|
| 1356 |
+
"epoch": 0.444,
|
| 1357 |
+
"grad_norm": 70.5,
|
| 1358 |
+
"learning_rate": 1.975777777777778e-05,
|
| 1359 |
+
"loss": 19.3826,
|
| 1360 |
+
"mean_token_accuracy": 0.6880253683775663,
|
| 1361 |
+
"num_tokens": 5414444.0,
|
| 1362 |
+
"step": 1110
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"entropy": 1.202802482806146,
|
| 1366 |
+
"epoch": 0.448,
|
| 1367 |
+
"grad_norm": 73.5,
|
| 1368 |
+
"learning_rate": 1.9735555555555556e-05,
|
| 1369 |
+
"loss": 19.4755,
|
| 1370 |
+
"mean_token_accuracy": 0.6911316126585006,
|
| 1371 |
+
"num_tokens": 5460559.0,
|
| 1372 |
+
"step": 1120
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"entropy": 1.1450790271162987,
|
| 1376 |
+
"epoch": 0.452,
|
| 1377 |
+
"grad_norm": 72.5,
|
| 1378 |
+
"learning_rate": 1.9713333333333337e-05,
|
| 1379 |
+
"loss": 18.4276,
|
| 1380 |
+
"mean_token_accuracy": 0.7056065019220114,
|
| 1381 |
+
"num_tokens": 5510112.0,
|
| 1382 |
+
"step": 1130
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"entropy": 1.2133123382925988,
|
| 1386 |
+
"epoch": 0.456,
|
| 1387 |
+
"grad_norm": 71.5,
|
| 1388 |
+
"learning_rate": 1.969111111111111e-05,
|
| 1389 |
+
"loss": 19.3288,
|
| 1390 |
+
"mean_token_accuracy": 0.6915038814768195,
|
| 1391 |
+
"num_tokens": 5558471.0,
|
| 1392 |
+
"step": 1140
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"entropy": 1.1987600516527892,
|
| 1396 |
+
"epoch": 0.46,
|
| 1397 |
+
"grad_norm": 73.5,
|
| 1398 |
+
"learning_rate": 1.9668888888888892e-05,
|
| 1399 |
+
"loss": 19.6063,
|
| 1400 |
+
"mean_token_accuracy": 0.684781968779862,
|
| 1401 |
+
"num_tokens": 5605237.0,
|
| 1402 |
+
"step": 1150
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"entropy": 1.2029788015410303,
|
| 1406 |
+
"epoch": 0.464,
|
| 1407 |
+
"grad_norm": 77.0,
|
| 1408 |
+
"learning_rate": 1.9646666666666666e-05,
|
| 1409 |
+
"loss": 19.113,
|
| 1410 |
+
"mean_token_accuracy": 0.6943607520312071,
|
| 1411 |
+
"num_tokens": 5653809.0,
|
| 1412 |
+
"step": 1160
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"entropy": 1.1871271941810846,
|
| 1416 |
+
"epoch": 0.468,
|
| 1417 |
+
"grad_norm": 80.0,
|
| 1418 |
+
"learning_rate": 1.9624444444444447e-05,
|
| 1419 |
+
"loss": 19.4077,
|
| 1420 |
+
"mean_token_accuracy": 0.6932028122246265,
|
| 1421 |
+
"num_tokens": 5699727.0,
|
| 1422 |
+
"step": 1170
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"entropy": 1.2316160146147013,
|
| 1426 |
+
"epoch": 0.472,
|
| 1427 |
+
"grad_norm": 67.0,
|
| 1428 |
+
"learning_rate": 1.9602222222222225e-05,
|
| 1429 |
+
"loss": 19.4937,
|
| 1430 |
+
"mean_token_accuracy": 0.6891589108854532,
|
| 1431 |
+
"num_tokens": 5751529.0,
|
| 1432 |
+
"step": 1180
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"entropy": 1.1813869891688227,
|
| 1436 |
+
"epoch": 0.476,
|
| 1437 |
+
"grad_norm": 82.5,
|
| 1438 |
+
"learning_rate": 1.9580000000000002e-05,
|
| 1439 |
+
"loss": 19.368,
|
| 1440 |
+
"mean_token_accuracy": 0.6905937325209379,
|
| 1441 |
+
"num_tokens": 5801167.0,
|
| 1442 |
+
"step": 1190
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"entropy": 1.2543733013793825,
|
| 1446 |
+
"epoch": 0.48,
|
| 1447 |
+
"grad_norm": 65.0,
|
| 1448 |
+
"learning_rate": 1.955777777777778e-05,
|
| 1449 |
+
"loss": 19.9054,
|
| 1450 |
+
"mean_token_accuracy": 0.6872919265180826,
|
| 1451 |
+
"num_tokens": 5850176.0,
|
| 1452 |
+
"step": 1200
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.48,
|
| 1456 |
+
"eval_biology_entropy": 1.2115116498470306,
|
| 1457 |
+
"eval_biology_loss": 1.1886382102966309,
|
| 1458 |
+
"eval_biology_mean_token_accuracy": 0.6968740136623383,
|
| 1459 |
+
"eval_biology_num_tokens": 5850176.0,
|
| 1460 |
+
"eval_biology_runtime": 34.2937,
|
| 1461 |
+
"eval_biology_samples_per_second": 14.58,
|
| 1462 |
+
"eval_biology_steps_per_second": 14.58,
|
| 1463 |
+
"step": 1200
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 0.48,
|
| 1467 |
+
"eval_chemistry_entropy": 0.7031495725512504,
|
| 1468 |
+
"eval_chemistry_loss": 0.6727164387702942,
|
| 1469 |
+
"eval_chemistry_mean_token_accuracy": 0.8113047035932541,
|
| 1470 |
+
"eval_chemistry_num_tokens": 5850176.0,
|
| 1471 |
+
"eval_chemistry_runtime": 39.0431,
|
| 1472 |
+
"eval_chemistry_samples_per_second": 12.806,
|
| 1473 |
+
"eval_chemistry_steps_per_second": 12.806,
|
| 1474 |
+
"step": 1200
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"entropy": 1.135433156043291,
|
| 1478 |
+
"epoch": 0.484,
|
| 1479 |
+
"grad_norm": 80.0,
|
| 1480 |
+
"learning_rate": 1.9535555555555557e-05,
|
| 1481 |
+
"loss": 18.5087,
|
| 1482 |
+
"mean_token_accuracy": 0.699444928765297,
|
| 1483 |
+
"num_tokens": 5897377.0,
|
| 1484 |
+
"step": 1210
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"entropy": 1.1549700409173966,
|
| 1488 |
+
"epoch": 0.488,
|
| 1489 |
+
"grad_norm": 84.5,
|
| 1490 |
+
"learning_rate": 1.9513333333333335e-05,
|
| 1491 |
+
"loss": 18.4892,
|
| 1492 |
+
"mean_token_accuracy": 0.701468913629651,
|
| 1493 |
+
"num_tokens": 5948898.0,
|
| 1494 |
+
"step": 1220
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"entropy": 1.154023634083569,
|
| 1498 |
+
"epoch": 0.492,
|
| 1499 |
+
"grad_norm": 80.5,
|
| 1500 |
+
"learning_rate": 1.9491111111111113e-05,
|
| 1501 |
+
"loss": 18.485,
|
| 1502 |
+
"mean_token_accuracy": 0.7046966258436441,
|
| 1503 |
+
"num_tokens": 5997563.0,
|
| 1504 |
+
"step": 1230
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"entropy": 1.167623796314001,
|
| 1508 |
+
"epoch": 0.496,
|
| 1509 |
+
"grad_norm": 81.0,
|
| 1510 |
+
"learning_rate": 1.946888888888889e-05,
|
| 1511 |
+
"loss": 18.7743,
|
| 1512 |
+
"mean_token_accuracy": 0.6952338460832834,
|
| 1513 |
+
"num_tokens": 6046503.0,
|
| 1514 |
+
"step": 1240
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"entropy": 1.2217689264565705,
|
| 1518 |
+
"epoch": 0.5,
|
| 1519 |
+
"grad_norm": 62.75,
|
| 1520 |
+
"learning_rate": 1.9446666666666668e-05,
|
| 1521 |
+
"loss": 19.5256,
|
| 1522 |
+
"mean_token_accuracy": 0.6905383057892323,
|
| 1523 |
+
"num_tokens": 6091587.0,
|
| 1524 |
+
"step": 1250
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"entropy": 1.1947737470269204,
|
| 1528 |
+
"epoch": 0.504,
|
| 1529 |
+
"grad_norm": 81.0,
|
| 1530 |
+
"learning_rate": 1.9424444444444446e-05,
|
| 1531 |
+
"loss": 19.5094,
|
| 1532 |
+
"mean_token_accuracy": 0.6902488011866807,
|
| 1533 |
+
"num_tokens": 6138649.0,
|
| 1534 |
+
"step": 1260
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"entropy": 1.1597765712067485,
|
| 1538 |
+
"epoch": 0.508,
|
| 1539 |
+
"grad_norm": 64.0,
|
| 1540 |
+
"learning_rate": 1.9402222222222223e-05,
|
| 1541 |
+
"loss": 18.4208,
|
| 1542 |
+
"mean_token_accuracy": 0.7028776211664081,
|
| 1543 |
+
"num_tokens": 6189122.0,
|
| 1544 |
+
"step": 1270
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"entropy": 1.2139740228652953,
|
| 1548 |
+
"epoch": 0.512,
|
| 1549 |
+
"grad_norm": 68.0,
|
| 1550 |
+
"learning_rate": 1.938e-05,
|
| 1551 |
+
"loss": 19.5968,
|
| 1552 |
+
"mean_token_accuracy": 0.6862009439617396,
|
| 1553 |
+
"num_tokens": 6240456.0,
|
| 1554 |
+
"step": 1280
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"entropy": 1.214512193016708,
|
| 1558 |
+
"epoch": 0.516,
|
| 1559 |
+
"grad_norm": 73.5,
|
| 1560 |
+
"learning_rate": 1.935777777777778e-05,
|
| 1561 |
+
"loss": 19.5979,
|
| 1562 |
+
"mean_token_accuracy": 0.6916845444589853,
|
| 1563 |
+
"num_tokens": 6286445.0,
|
| 1564 |
+
"step": 1290
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"entropy": 1.177449732273817,
|
| 1568 |
+
"epoch": 0.52,
|
| 1569 |
+
"grad_norm": 72.5,
|
| 1570 |
+
"learning_rate": 1.9335555555555556e-05,
|
| 1571 |
+
"loss": 19.0899,
|
| 1572 |
+
"mean_token_accuracy": 0.6944207726046443,
|
| 1573 |
+
"num_tokens": 6333958.0,
|
| 1574 |
+
"step": 1300
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 0.52,
|
| 1578 |
+
"eval_biology_entropy": 1.1935186836719514,
|
| 1579 |
+
"eval_biology_loss": 1.1775966882705688,
|
| 1580 |
+
"eval_biology_mean_token_accuracy": 0.6992735857367516,
|
| 1581 |
+
"eval_biology_num_tokens": 6333958.0,
|
| 1582 |
+
"eval_biology_runtime": 34.2425,
|
| 1583 |
+
"eval_biology_samples_per_second": 14.602,
|
| 1584 |
+
"eval_biology_steps_per_second": 14.602,
|
| 1585 |
+
"step": 1300
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.52,
|
| 1589 |
+
"eval_chemistry_entropy": 0.6942039486765862,
|
| 1590 |
+
"eval_chemistry_loss": 0.6730213165283203,
|
| 1591 |
+
"eval_chemistry_mean_token_accuracy": 0.8110301185846329,
|
| 1592 |
+
"eval_chemistry_num_tokens": 6333958.0,
|
| 1593 |
+
"eval_chemistry_runtime": 39.0845,
|
| 1594 |
+
"eval_chemistry_samples_per_second": 12.793,
|
| 1595 |
+
"eval_chemistry_steps_per_second": 12.793,
|
| 1596 |
+
"step": 1300
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"entropy": 1.182255671173334,
|
| 1600 |
+
"epoch": 0.524,
|
| 1601 |
+
"grad_norm": 87.0,
|
| 1602 |
+
"learning_rate": 1.9313333333333334e-05,
|
| 1603 |
+
"loss": 19.022,
|
| 1604 |
+
"mean_token_accuracy": 0.6953186947852373,
|
| 1605 |
+
"num_tokens": 6383679.0,
|
| 1606 |
+
"step": 1310
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"entropy": 1.225791597366333,
|
| 1610 |
+
"epoch": 0.528,
|
| 1611 |
+
"grad_norm": 66.0,
|
| 1612 |
+
"learning_rate": 1.9291111111111115e-05,
|
| 1613 |
+
"loss": 19.8175,
|
| 1614 |
+
"mean_token_accuracy": 0.6822869379073382,
|
| 1615 |
+
"num_tokens": 6430555.0,
|
| 1616 |
+
"step": 1320
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"entropy": 1.1287783520296215,
|
| 1620 |
+
"epoch": 0.532,
|
| 1621 |
+
"grad_norm": 82.5,
|
| 1622 |
+
"learning_rate": 1.926888888888889e-05,
|
| 1623 |
+
"loss": 18.257,
|
| 1624 |
+
"mean_token_accuracy": 0.7016044341027736,
|
| 1625 |
+
"num_tokens": 6481759.0,
|
| 1626 |
+
"step": 1330
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"entropy": 1.229908336699009,
|
| 1630 |
+
"epoch": 0.536,
|
| 1631 |
+
"grad_norm": 72.0,
|
| 1632 |
+
"learning_rate": 1.924666666666667e-05,
|
| 1633 |
+
"loss": 19.7004,
|
| 1634 |
+
"mean_token_accuracy": 0.6902735695242882,
|
| 1635 |
+
"num_tokens": 6533033.0,
|
| 1636 |
+
"step": 1340
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"entropy": 1.1814987678080797,
|
| 1640 |
+
"epoch": 0.54,
|
| 1641 |
+
"grad_norm": 78.5,
|
| 1642 |
+
"learning_rate": 1.9224444444444444e-05,
|
| 1643 |
+
"loss": 18.9379,
|
| 1644 |
+
"mean_token_accuracy": 0.699426869675517,
|
| 1645 |
+
"num_tokens": 6580914.0,
|
| 1646 |
+
"step": 1350
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"entropy": 1.1593723269179463,
|
| 1650 |
+
"epoch": 0.544,
|
| 1651 |
+
"grad_norm": 67.5,
|
| 1652 |
+
"learning_rate": 1.9202222222222225e-05,
|
| 1653 |
+
"loss": 18.785,
|
| 1654 |
+
"mean_token_accuracy": 0.6992750339210033,
|
| 1655 |
+
"num_tokens": 6626006.0,
|
| 1656 |
+
"step": 1360
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"entropy": 1.1956233188509942,
|
| 1660 |
+
"epoch": 0.548,
|
| 1661 |
+
"grad_norm": 75.0,
|
| 1662 |
+
"learning_rate": 1.918e-05,
|
| 1663 |
+
"loss": 19.3378,
|
| 1664 |
+
"mean_token_accuracy": 0.6945411875844002,
|
| 1665 |
+
"num_tokens": 6672679.0,
|
| 1666 |
+
"step": 1370
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"entropy": 1.2158267103135585,
|
| 1670 |
+
"epoch": 0.552,
|
| 1671 |
+
"grad_norm": 74.5,
|
| 1672 |
+
"learning_rate": 1.915777777777778e-05,
|
| 1673 |
+
"loss": 19.4937,
|
| 1674 |
+
"mean_token_accuracy": 0.6895309235900641,
|
| 1675 |
+
"num_tokens": 6719864.0,
|
| 1676 |
+
"step": 1380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"entropy": 1.1711071468889713,
|
| 1680 |
+
"epoch": 0.556,
|
| 1681 |
+
"grad_norm": 70.0,
|
| 1682 |
+
"learning_rate": 1.9135555555555555e-05,
|
| 1683 |
+
"loss": 18.9627,
|
| 1684 |
+
"mean_token_accuracy": 0.6955475777387619,
|
| 1685 |
+
"num_tokens": 6768288.0,
|
| 1686 |
+
"step": 1390
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"entropy": 1.1587215423583985,
|
| 1690 |
+
"epoch": 0.56,
|
| 1691 |
+
"grad_norm": 70.0,
|
| 1692 |
+
"learning_rate": 1.9113333333333336e-05,
|
| 1693 |
+
"loss": 18.6741,
|
| 1694 |
+
"mean_token_accuracy": 0.6977546285837889,
|
| 1695 |
+
"num_tokens": 6820754.0,
|
| 1696 |
+
"step": 1400
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.56,
|
| 1700 |
+
"eval_biology_entropy": 1.1666600781083107,
|
| 1701 |
+
"eval_biology_loss": 1.1740920543670654,
|
| 1702 |
+
"eval_biology_mean_token_accuracy": 0.7002735708355904,
|
| 1703 |
+
"eval_biology_num_tokens": 6820754.0,
|
| 1704 |
+
"eval_biology_runtime": 34.4305,
|
| 1705 |
+
"eval_biology_samples_per_second": 14.522,
|
| 1706 |
+
"eval_biology_steps_per_second": 14.522,
|
| 1707 |
+
"step": 1400
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 0.56,
|
| 1711 |
+
"eval_chemistry_entropy": 0.6812023703157902,
|
| 1712 |
+
"eval_chemistry_loss": 0.6732838749885559,
|
| 1713 |
+
"eval_chemistry_mean_token_accuracy": 0.8108260444402695,
|
| 1714 |
+
"eval_chemistry_num_tokens": 6820754.0,
|
| 1715 |
+
"eval_chemistry_runtime": 39.1375,
|
| 1716 |
+
"eval_chemistry_samples_per_second": 12.775,
|
| 1717 |
+
"eval_chemistry_steps_per_second": 12.775,
|
| 1718 |
+
"step": 1400
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"entropy": 1.1611128127202393,
|
| 1722 |
+
"epoch": 0.564,
|
| 1723 |
+
"grad_norm": 69.0,
|
| 1724 |
+
"learning_rate": 1.9091111111111113e-05,
|
| 1725 |
+
"loss": 18.4737,
|
| 1726 |
+
"mean_token_accuracy": 0.7005460750311613,
|
| 1727 |
+
"num_tokens": 6875015.0,
|
| 1728 |
+
"step": 1410
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"entropy": 1.1703616185113788,
|
| 1732 |
+
"epoch": 0.568,
|
| 1733 |
+
"grad_norm": 55.0,
|
| 1734 |
+
"learning_rate": 1.906888888888889e-05,
|
| 1735 |
+
"loss": 18.9916,
|
| 1736 |
+
"mean_token_accuracy": 0.6964723881334066,
|
| 1737 |
+
"num_tokens": 6924942.0,
|
| 1738 |
+
"step": 1420
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"entropy": 1.2203245889395475,
|
| 1742 |
+
"epoch": 0.572,
|
| 1743 |
+
"grad_norm": 69.5,
|
| 1744 |
+
"learning_rate": 1.904666666666667e-05,
|
| 1745 |
+
"loss": 19.8421,
|
| 1746 |
+
"mean_token_accuracy": 0.6847291864454746,
|
| 1747 |
+
"num_tokens": 6969801.0,
|
| 1748 |
+
"step": 1430
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"entropy": 1.1570590244606138,
|
| 1752 |
+
"epoch": 0.576,
|
| 1753 |
+
"grad_norm": 79.5,
|
| 1754 |
+
"learning_rate": 1.9024444444444446e-05,
|
| 1755 |
+
"loss": 18.3389,
|
| 1756 |
+
"mean_token_accuracy": 0.7031517468392849,
|
| 1757 |
+
"num_tokens": 7021844.0,
|
| 1758 |
+
"step": 1440
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"entropy": 1.1714151045307517,
|
| 1762 |
+
"epoch": 0.58,
|
| 1763 |
+
"grad_norm": 78.0,
|
| 1764 |
+
"learning_rate": 1.9002222222222224e-05,
|
| 1765 |
+
"loss": 19.0768,
|
| 1766 |
+
"mean_token_accuracy": 0.695703462511301,
|
| 1767 |
+
"num_tokens": 7070185.0,
|
| 1768 |
+
"step": 1450
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"entropy": 1.1976398654282092,
|
| 1772 |
+
"epoch": 0.584,
|
| 1773 |
+
"grad_norm": 68.5,
|
| 1774 |
+
"learning_rate": 1.898e-05,
|
| 1775 |
+
"loss": 19.1845,
|
| 1776 |
+
"mean_token_accuracy": 0.6903858676552772,
|
| 1777 |
+
"num_tokens": 7118443.0,
|
| 1778 |
+
"step": 1460
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"entropy": 1.1526124764233827,
|
| 1782 |
+
"epoch": 0.588,
|
| 1783 |
+
"grad_norm": 79.0,
|
| 1784 |
+
"learning_rate": 1.895777777777778e-05,
|
| 1785 |
+
"loss": 18.7328,
|
| 1786 |
+
"mean_token_accuracy": 0.6993513517081738,
|
| 1787 |
+
"num_tokens": 7165265.0,
|
| 1788 |
+
"step": 1470
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"entropy": 1.152988604642451,
|
| 1792 |
+
"epoch": 0.592,
|
| 1793 |
+
"grad_norm": 68.0,
|
| 1794 |
+
"learning_rate": 1.8935555555555556e-05,
|
| 1795 |
+
"loss": 18.4211,
|
| 1796 |
+
"mean_token_accuracy": 0.7030052304267883,
|
| 1797 |
+
"num_tokens": 7213951.0,
|
| 1798 |
+
"step": 1480
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"entropy": 1.1504094721749425,
|
| 1802 |
+
"epoch": 0.596,
|
| 1803 |
+
"grad_norm": 72.5,
|
| 1804 |
+
"learning_rate": 1.8913333333333334e-05,
|
| 1805 |
+
"loss": 18.6603,
|
| 1806 |
+
"mean_token_accuracy": 0.6974705925211311,
|
| 1807 |
+
"num_tokens": 7265612.0,
|
| 1808 |
+
"step": 1490
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"entropy": 1.1832087337970734,
|
| 1812 |
+
"epoch": 0.6,
|
| 1813 |
+
"grad_norm": 82.0,
|
| 1814 |
+
"learning_rate": 1.8891111111111115e-05,
|
| 1815 |
+
"loss": 18.9664,
|
| 1816 |
+
"mean_token_accuracy": 0.6971027232706547,
|
| 1817 |
+
"num_tokens": 7317984.0,
|
| 1818 |
+
"step": 1500
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 0.6,
|
| 1822 |
+
"eval_biology_entropy": 1.1557998611927032,
|
| 1823 |
+
"eval_biology_loss": 1.1672067642211914,
|
| 1824 |
+
"eval_biology_mean_token_accuracy": 0.7016319183707237,
|
| 1825 |
+
"eval_biology_num_tokens": 7317984.0,
|
| 1826 |
+
"eval_biology_runtime": 34.2829,
|
| 1827 |
+
"eval_biology_samples_per_second": 14.585,
|
| 1828 |
+
"eval_biology_steps_per_second": 14.585,
|
| 1829 |
+
"step": 1500
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.6,
|
| 1833 |
+
"eval_chemistry_entropy": 0.6806726072132587,
|
| 1834 |
+
"eval_chemistry_loss": 0.670280396938324,
|
| 1835 |
+
"eval_chemistry_mean_token_accuracy": 0.8115782773494721,
|
| 1836 |
+
"eval_chemistry_num_tokens": 7317984.0,
|
| 1837 |
+
"eval_chemistry_runtime": 39.0952,
|
| 1838 |
+
"eval_chemistry_samples_per_second": 12.789,
|
| 1839 |
+
"eval_chemistry_steps_per_second": 12.789,
|
| 1840 |
+
"step": 1500
|
| 1841 |
+
}
|
| 1842 |
+
],
|
| 1843 |
+
"logging_steps": 10,
|
| 1844 |
+
"max_steps": 10000,
|
| 1845 |
+
"num_input_tokens_seen": 0,
|
| 1846 |
+
"num_train_epochs": 4,
|
| 1847 |
+
"save_steps": 500,
|
| 1848 |
+
"stateful_callbacks": {
|
| 1849 |
+
"TrainerControl": {
|
| 1850 |
+
"args": {
|
| 1851 |
+
"should_epoch_stop": false,
|
| 1852 |
+
"should_evaluate": false,
|
| 1853 |
+
"should_log": false,
|
| 1854 |
+
"should_save": true,
|
| 1855 |
+
"should_training_stop": false
|
| 1856 |
+
},
|
| 1857 |
+
"attributes": {}
|
| 1858 |
+
}
|
| 1859 |
+
},
|
| 1860 |
+
"total_flos": 4.90702780211162e+17,
|
| 1861 |
+
"train_batch_size": 1,
|
| 1862 |
+
"trial_name": null,
|
| 1863 |
+
"trial_params": null
|
| 1864 |
+
}
|
checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:80b5712b7c29bfc74fdd744346ebce13b8126ec518d448ca496e5a80dda84b00
|
| 3 |
+
size 6481
|