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 +2194 -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 sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
|
| 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 sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
|
| 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 sha256:1c5464143ebdecf512562ea3287caadec223e9c8dd2eb85a6ac6316f42ede61a
|
| 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 sha256:8995d088ea034e0055ffdee2e1b467fb0ce21e8e37818703b02a15d76da5cce9
|
| 3 |
+
size 3187869259
|
checkpoint-1500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2fc8057a0ec2b2e31b842037349765b4236ef308dc0c5dde871c8370ee3eff6b
|
| 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 +1,2194 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.6676681689200468,
|
| 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": 0.548704273160547,
|
| 14 |
+
"epoch": 0.004451121126133645,
|
| 15 |
+
"grad_norm": 1192.0,
|
| 16 |
+
"learning_rate": 1.8e-07,
|
| 17 |
+
"loss": 47.4036,
|
| 18 |
+
"mean_token_accuracy": 0.641148860566318,
|
| 19 |
+
"num_tokens": 63133.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 0.534384406823665,
|
| 24 |
+
"epoch": 0.00890224225226729,
|
| 25 |
+
"grad_norm": 1328.0,
|
| 26 |
+
"learning_rate": 3.8e-07,
|
| 27 |
+
"loss": 50.4685,
|
| 28 |
+
"mean_token_accuracy": 0.6258877433836461,
|
| 29 |
+
"num_tokens": 128968.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 0.537126874178648,
|
| 34 |
+
"epoch": 0.013353363378400934,
|
| 35 |
+
"grad_norm": 1320.0,
|
| 36 |
+
"learning_rate": 5.800000000000001e-07,
|
| 37 |
+
"loss": 49.4492,
|
| 38 |
+
"mean_token_accuracy": 0.6289473636075854,
|
| 39 |
+
"num_tokens": 193314.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 0.5729353995062411,
|
| 44 |
+
"epoch": 0.01780448450453458,
|
| 45 |
+
"grad_norm": 1012.0,
|
| 46 |
+
"learning_rate": 7.8e-07,
|
| 47 |
+
"loss": 49.5744,
|
| 48 |
+
"mean_token_accuracy": 0.6247817112132907,
|
| 49 |
+
"num_tokens": 257189.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 0.5564130786806345,
|
| 54 |
+
"epoch": 0.022255605630668224,
|
| 55 |
+
"grad_norm": 1096.0,
|
| 56 |
+
"learning_rate": 9.800000000000001e-07,
|
| 57 |
+
"loss": 44.4523,
|
| 58 |
+
"mean_token_accuracy": 0.6425775093957782,
|
| 59 |
+
"num_tokens": 324224.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 0.590752430446446,
|
| 64 |
+
"epoch": 0.026706726756801868,
|
| 65 |
+
"grad_norm": 816.0,
|
| 66 |
+
"learning_rate": 1.1800000000000001e-06,
|
| 67 |
+
"loss": 43.7528,
|
| 68 |
+
"mean_token_accuracy": 0.6364219607785344,
|
| 69 |
+
"num_tokens": 394654.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 0.6452584797516465,
|
| 74 |
+
"epoch": 0.031157847882935515,
|
| 75 |
+
"grad_norm": 600.0,
|
| 76 |
+
"learning_rate": 1.3800000000000001e-06,
|
| 77 |
+
"loss": 39.6529,
|
| 78 |
+
"mean_token_accuracy": 0.6429168622940779,
|
| 79 |
+
"num_tokens": 461658.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 0.7016998372972012,
|
| 84 |
+
"epoch": 0.03560896900906916,
|
| 85 |
+
"grad_norm": 450.0,
|
| 86 |
+
"learning_rate": 1.5800000000000001e-06,
|
| 87 |
+
"loss": 35.508,
|
| 88 |
+
"mean_token_accuracy": 0.6540533849969506,
|
| 89 |
+
"num_tokens": 528285.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 0.8241214359179139,
|
| 94 |
+
"epoch": 0.04006009013520281,
|
| 95 |
+
"grad_norm": 320.0,
|
| 96 |
+
"learning_rate": 1.7800000000000001e-06,
|
| 97 |
+
"loss": 34.6569,
|
| 98 |
+
"mean_token_accuracy": 0.6461574586108327,
|
| 99 |
+
"num_tokens": 595336.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.0065209370106458,
|
| 104 |
+
"epoch": 0.04451121126133645,
|
| 105 |
+
"grad_norm": 208.0,
|
| 106 |
+
"learning_rate": 1.98e-06,
|
| 107 |
+
"loss": 31.2419,
|
| 108 |
+
"mean_token_accuracy": 0.6523716723546386,
|
| 109 |
+
"num_tokens": 663126.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.04451121126133645,
|
| 114 |
+
"eval_biology_entropy": 1.5740653836727143,
|
| 115 |
+
"eval_biology_loss": 2.817919969558716,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.5265991420149804,
|
| 117 |
+
"eval_biology_num_tokens": 663126.0,
|
| 118 |
+
"eval_biology_runtime": 49.0691,
|
| 119 |
+
"eval_biology_samples_per_second": 10.19,
|
| 120 |
+
"eval_biology_steps_per_second": 10.19,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04451121126133645,
|
| 125 |
+
"eval_chemistry_entropy": 1.1085504465103149,
|
| 126 |
+
"eval_chemistry_loss": 1.8567637205123901,
|
| 127 |
+
"eval_chemistry_mean_token_accuracy": 0.6607221029996871,
|
| 128 |
+
"eval_chemistry_num_tokens": 663126.0,
|
| 129 |
+
"eval_chemistry_runtime": 56.0546,
|
| 130 |
+
"eval_chemistry_samples_per_second": 8.92,
|
| 131 |
+
"eval_chemistry_steps_per_second": 8.92,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.04451121126133645,
|
| 136 |
+
"eval_math_entropy": 0.7685319324731826,
|
| 137 |
+
"eval_math_loss": 1.6490864753723145,
|
| 138 |
+
"eval_math_mean_token_accuracy": 0.701621960401535,
|
| 139 |
+
"eval_math_num_tokens": 663126.0,
|
| 140 |
+
"eval_math_runtime": 57.7457,
|
| 141 |
+
"eval_math_samples_per_second": 8.659,
|
| 142 |
+
"eval_math_steps_per_second": 8.659,
|
| 143 |
+
"step": 100
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.04451121126133645,
|
| 147 |
+
"eval_physics_entropy": 0.9800559472441673,
|
| 148 |
+
"eval_physics_loss": 1.658416986465454,
|
| 149 |
+
"eval_physics_mean_token_accuracy": 0.6876975857019424,
|
| 150 |
+
"eval_physics_num_tokens": 663126.0,
|
| 151 |
+
"eval_physics_runtime": 64.5823,
|
| 152 |
+
"eval_physics_samples_per_second": 7.742,
|
| 153 |
+
"eval_physics_steps_per_second": 7.742,
|
| 154 |
+
"step": 100
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"entropy": 1.2310039116069675,
|
| 158 |
+
"epoch": 0.048962332387470095,
|
| 159 |
+
"grad_norm": 129.0,
|
| 160 |
+
"learning_rate": 2.1800000000000003e-06,
|
| 161 |
+
"loss": 28.6754,
|
| 162 |
+
"mean_token_accuracy": 0.6632729232311249,
|
| 163 |
+
"num_tokens": 731060.0,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"entropy": 1.4431365672498941,
|
| 168 |
+
"epoch": 0.053413453513603736,
|
| 169 |
+
"grad_norm": 86.0,
|
| 170 |
+
"learning_rate": 2.38e-06,
|
| 171 |
+
"loss": 26.3421,
|
| 172 |
+
"mean_token_accuracy": 0.6755226453766227,
|
| 173 |
+
"num_tokens": 801635.0,
|
| 174 |
+
"step": 120
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"entropy": 1.4806174896657467,
|
| 178 |
+
"epoch": 0.05786457463973738,
|
| 179 |
+
"grad_norm": 72.0,
|
| 180 |
+
"learning_rate": 2.5800000000000003e-06,
|
| 181 |
+
"loss": 25.6649,
|
| 182 |
+
"mean_token_accuracy": 0.6807831462472678,
|
| 183 |
+
"num_tokens": 867260.0,
|
| 184 |
+
"step": 130
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"entropy": 1.511351894028485,
|
| 188 |
+
"epoch": 0.06231569576587103,
|
| 189 |
+
"grad_norm": 95.5,
|
| 190 |
+
"learning_rate": 2.7800000000000005e-06,
|
| 191 |
+
"loss": 25.1648,
|
| 192 |
+
"mean_token_accuracy": 0.6787236794829369,
|
| 193 |
+
"num_tokens": 931344.0,
|
| 194 |
+
"step": 140
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"entropy": 1.4235187267884613,
|
| 198 |
+
"epoch": 0.06676681689200467,
|
| 199 |
+
"grad_norm": 57.25,
|
| 200 |
+
"learning_rate": 2.9800000000000003e-06,
|
| 201 |
+
"loss": 23.0409,
|
| 202 |
+
"mean_token_accuracy": 0.6943066673353314,
|
| 203 |
+
"num_tokens": 993586.0,
|
| 204 |
+
"step": 150
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"entropy": 1.36321877297014,
|
| 208 |
+
"epoch": 0.07121793801813832,
|
| 209 |
+
"grad_norm": 52.75,
|
| 210 |
+
"learning_rate": 3.1800000000000005e-06,
|
| 211 |
+
"loss": 22.2695,
|
| 212 |
+
"mean_token_accuracy": 0.6991008354350925,
|
| 213 |
+
"num_tokens": 1060812.0,
|
| 214 |
+
"step": 160
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"entropy": 1.2639451341703534,
|
| 218 |
+
"epoch": 0.07566905914427197,
|
| 219 |
+
"grad_norm": 55.0,
|
| 220 |
+
"learning_rate": 3.3800000000000007e-06,
|
| 221 |
+
"loss": 20.5103,
|
| 222 |
+
"mean_token_accuracy": 0.7160237554460764,
|
| 223 |
+
"num_tokens": 1131832.0,
|
| 224 |
+
"step": 170
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"entropy": 1.3122306695207953,
|
| 228 |
+
"epoch": 0.08012018027040561,
|
| 229 |
+
"grad_norm": 43.25,
|
| 230 |
+
"learning_rate": 3.58e-06,
|
| 231 |
+
"loss": 20.6467,
|
| 232 |
+
"mean_token_accuracy": 0.7132720120251179,
|
| 233 |
+
"num_tokens": 1200650.0,
|
| 234 |
+
"step": 180
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"entropy": 1.321925420127809,
|
| 238 |
+
"epoch": 0.08457130139653925,
|
| 239 |
+
"grad_norm": 52.25,
|
| 240 |
+
"learning_rate": 3.7800000000000002e-06,
|
| 241 |
+
"loss": 20.6542,
|
| 242 |
+
"mean_token_accuracy": 0.7144606521353125,
|
| 243 |
+
"num_tokens": 1265107.0,
|
| 244 |
+
"step": 190
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"entropy": 1.304779276624322,
|
| 248 |
+
"epoch": 0.0890224225226729,
|
| 249 |
+
"grad_norm": 51.75,
|
| 250 |
+
"learning_rate": 3.980000000000001e-06,
|
| 251 |
+
"loss": 19.8821,
|
| 252 |
+
"mean_token_accuracy": 0.7192140439525246,
|
| 253 |
+
"num_tokens": 1327380.0,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.0890224225226729,
|
| 258 |
+
"eval_biology_entropy": 1.7584778207540512,
|
| 259 |
+
"eval_biology_loss": 2.0029563903808594,
|
| 260 |
+
"eval_biology_mean_token_accuracy": 0.5894142254590988,
|
| 261 |
+
"eval_biology_num_tokens": 1327380.0,
|
| 262 |
+
"eval_biology_runtime": 48.4623,
|
| 263 |
+
"eval_biology_samples_per_second": 10.317,
|
| 264 |
+
"eval_biology_steps_per_second": 10.317,
|
| 265 |
+
"step": 200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.0890224225226729,
|
| 269 |
+
"eval_chemistry_entropy": 1.2313024162650108,
|
| 270 |
+
"eval_chemistry_loss": 1.23660409450531,
|
| 271 |
+
"eval_chemistry_mean_token_accuracy": 0.7191956076025963,
|
| 272 |
+
"eval_chemistry_num_tokens": 1327380.0,
|
| 273 |
+
"eval_chemistry_runtime": 56.1215,
|
| 274 |
+
"eval_chemistry_samples_per_second": 8.909,
|
| 275 |
+
"eval_chemistry_steps_per_second": 8.909,
|
| 276 |
+
"step": 200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.0890224225226729,
|
| 280 |
+
"eval_math_entropy": 0.9544051561951638,
|
| 281 |
+
"eval_math_loss": 1.1321452856063843,
|
| 282 |
+
"eval_math_mean_token_accuracy": 0.745829626917839,
|
| 283 |
+
"eval_math_num_tokens": 1327380.0,
|
| 284 |
+
"eval_math_runtime": 57.6893,
|
| 285 |
+
"eval_math_samples_per_second": 8.667,
|
| 286 |
+
"eval_math_steps_per_second": 8.667,
|
| 287 |
+
"step": 200
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.0890224225226729,
|
| 291 |
+
"eval_physics_entropy": 1.097377147436142,
|
| 292 |
+
"eval_physics_loss": 1.1237763166427612,
|
| 293 |
+
"eval_physics_mean_token_accuracy": 0.7436373601555825,
|
| 294 |
+
"eval_physics_num_tokens": 1327380.0,
|
| 295 |
+
"eval_physics_runtime": 64.6254,
|
| 296 |
+
"eval_physics_samples_per_second": 7.737,
|
| 297 |
+
"eval_physics_steps_per_second": 7.737,
|
| 298 |
+
"step": 200
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"entropy": 1.2635900884866715,
|
| 302 |
+
"epoch": 0.09347354364880654,
|
| 303 |
+
"grad_norm": 54.5,
|
| 304 |
+
"learning_rate": 4.18e-06,
|
| 305 |
+
"loss": 20.5549,
|
| 306 |
+
"mean_token_accuracy": 0.7137404002249241,
|
| 307 |
+
"num_tokens": 1393379.0,
|
| 308 |
+
"step": 210
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"entropy": 1.253603531420231,
|
| 312 |
+
"epoch": 0.09792466477494019,
|
| 313 |
+
"grad_norm": 45.25,
|
| 314 |
+
"learning_rate": 4.38e-06,
|
| 315 |
+
"loss": 19.3171,
|
| 316 |
+
"mean_token_accuracy": 0.7269632073119283,
|
| 317 |
+
"num_tokens": 1460130.0,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"entropy": 1.205259626172483,
|
| 322 |
+
"epoch": 0.10237578590107384,
|
| 323 |
+
"grad_norm": 50.5,
|
| 324 |
+
"learning_rate": 4.58e-06,
|
| 325 |
+
"loss": 18.3853,
|
| 326 |
+
"mean_token_accuracy": 0.7334358854219317,
|
| 327 |
+
"num_tokens": 1529183.0,
|
| 328 |
+
"step": 230
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"entropy": 1.2351684581488371,
|
| 332 |
+
"epoch": 0.10682690702720747,
|
| 333 |
+
"grad_norm": 61.5,
|
| 334 |
+
"learning_rate": 4.78e-06,
|
| 335 |
+
"loss": 19.0263,
|
| 336 |
+
"mean_token_accuracy": 0.723250618390739,
|
| 337 |
+
"num_tokens": 1597405.0,
|
| 338 |
+
"step": 240
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"entropy": 1.2351113758981227,
|
| 342 |
+
"epoch": 0.11127802815334112,
|
| 343 |
+
"grad_norm": 43.0,
|
| 344 |
+
"learning_rate": 4.980000000000001e-06,
|
| 345 |
+
"loss": 18.9974,
|
| 346 |
+
"mean_token_accuracy": 0.7199854323640466,
|
| 347 |
+
"num_tokens": 1661767.0,
|
| 348 |
+
"step": 250
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"entropy": 1.1816746497526764,
|
| 352 |
+
"epoch": 0.11572914927947477,
|
| 353 |
+
"grad_norm": 45.75,
|
| 354 |
+
"learning_rate": 5.18e-06,
|
| 355 |
+
"loss": 18.6797,
|
| 356 |
+
"mean_token_accuracy": 0.7243335520848632,
|
| 357 |
+
"num_tokens": 1728207.0,
|
| 358 |
+
"step": 260
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"entropy": 1.1266852594912051,
|
| 362 |
+
"epoch": 0.12018027040560841,
|
| 363 |
+
"grad_norm": 45.75,
|
| 364 |
+
"learning_rate": 5.380000000000001e-06,
|
| 365 |
+
"loss": 17.8377,
|
| 366 |
+
"mean_token_accuracy": 0.732505951821804,
|
| 367 |
+
"num_tokens": 1796055.0,
|
| 368 |
+
"step": 270
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"entropy": 1.1155648654326797,
|
| 372 |
+
"epoch": 0.12463139153174206,
|
| 373 |
+
"grad_norm": 49.0,
|
| 374 |
+
"learning_rate": 5.580000000000001e-06,
|
| 375 |
+
"loss": 17.5397,
|
| 376 |
+
"mean_token_accuracy": 0.7362749936059118,
|
| 377 |
+
"num_tokens": 1859684.0,
|
| 378 |
+
"step": 280
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"entropy": 1.1514084933325648,
|
| 382 |
+
"epoch": 0.1290825126578757,
|
| 383 |
+
"grad_norm": 51.0,
|
| 384 |
+
"learning_rate": 5.78e-06,
|
| 385 |
+
"loss": 18.5451,
|
| 386 |
+
"mean_token_accuracy": 0.723321282491088,
|
| 387 |
+
"num_tokens": 1921704.0,
|
| 388 |
+
"step": 290
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"entropy": 1.1451087817549706,
|
| 392 |
+
"epoch": 0.13353363378400934,
|
| 393 |
+
"grad_norm": 42.5,
|
| 394 |
+
"learning_rate": 5.98e-06,
|
| 395 |
+
"loss": 18.2063,
|
| 396 |
+
"mean_token_accuracy": 0.7288043411448598,
|
| 397 |
+
"num_tokens": 1985766.0,
|
| 398 |
+
"step": 300
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.13353363378400934,
|
| 402 |
+
"eval_biology_entropy": 1.5927132781147957,
|
| 403 |
+
"eval_biology_loss": 1.7975378036499023,
|
| 404 |
+
"eval_biology_mean_token_accuracy": 0.6119548050761223,
|
| 405 |
+
"eval_biology_num_tokens": 1985766.0,
|
| 406 |
+
"eval_biology_runtime": 48.6056,
|
| 407 |
+
"eval_biology_samples_per_second": 10.287,
|
| 408 |
+
"eval_biology_steps_per_second": 10.287,
|
| 409 |
+
"step": 300
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.13353363378400934,
|
| 413 |
+
"eval_chemistry_entropy": 1.0956866154670715,
|
| 414 |
+
"eval_chemistry_loss": 1.0781587362289429,
|
| 415 |
+
"eval_chemistry_mean_token_accuracy": 0.7388799833655357,
|
| 416 |
+
"eval_chemistry_num_tokens": 1985766.0,
|
| 417 |
+
"eval_chemistry_runtime": 55.9983,
|
| 418 |
+
"eval_chemistry_samples_per_second": 8.929,
|
| 419 |
+
"eval_chemistry_steps_per_second": 8.929,
|
| 420 |
+
"step": 300
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.13353363378400934,
|
| 424 |
+
"eval_math_entropy": 0.8625128073692322,
|
| 425 |
+
"eval_math_loss": 1.0247634649276733,
|
| 426 |
+
"eval_math_mean_token_accuracy": 0.7569631568193436,
|
| 427 |
+
"eval_math_num_tokens": 1985766.0,
|
| 428 |
+
"eval_math_runtime": 57.7379,
|
| 429 |
+
"eval_math_samples_per_second": 8.66,
|
| 430 |
+
"eval_math_steps_per_second": 8.66,
|
| 431 |
+
"step": 300
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.13353363378400934,
|
| 435 |
+
"eval_physics_entropy": 0.9960197100043297,
|
| 436 |
+
"eval_physics_loss": 0.993407666683197,
|
| 437 |
+
"eval_physics_mean_token_accuracy": 0.7602200539112091,
|
| 438 |
+
"eval_physics_num_tokens": 1985766.0,
|
| 439 |
+
"eval_physics_runtime": 64.6649,
|
| 440 |
+
"eval_physics_samples_per_second": 7.732,
|
| 441 |
+
"eval_physics_steps_per_second": 7.732,
|
| 442 |
+
"step": 300
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"entropy": 1.1051979778334498,
|
| 446 |
+
"epoch": 0.137984754910143,
|
| 447 |
+
"grad_norm": 49.5,
|
| 448 |
+
"learning_rate": 6.18e-06,
|
| 449 |
+
"loss": 17.5537,
|
| 450 |
+
"mean_token_accuracy": 0.7314090682193637,
|
| 451 |
+
"num_tokens": 2049041.0,
|
| 452 |
+
"step": 310
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"entropy": 1.0414688095450402,
|
| 456 |
+
"epoch": 0.14243587603627664,
|
| 457 |
+
"grad_norm": 37.75,
|
| 458 |
+
"learning_rate": 6.380000000000001e-06,
|
| 459 |
+
"loss": 16.669,
|
| 460 |
+
"mean_token_accuracy": 0.7476102706044913,
|
| 461 |
+
"num_tokens": 2118068.0,
|
| 462 |
+
"step": 320
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"entropy": 1.030502800643444,
|
| 466 |
+
"epoch": 0.14688699716241027,
|
| 467 |
+
"grad_norm": 46.5,
|
| 468 |
+
"learning_rate": 6.5800000000000005e-06,
|
| 469 |
+
"loss": 16.4295,
|
| 470 |
+
"mean_token_accuracy": 0.7489048192277551,
|
| 471 |
+
"num_tokens": 2186387.0,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"entropy": 1.0635908702388406,
|
| 476 |
+
"epoch": 0.15133811828854393,
|
| 477 |
+
"grad_norm": 42.0,
|
| 478 |
+
"learning_rate": 6.780000000000001e-06,
|
| 479 |
+
"loss": 17.235,
|
| 480 |
+
"mean_token_accuracy": 0.7380516286939383,
|
| 481 |
+
"num_tokens": 2252650.0,
|
| 482 |
+
"step": 340
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"entropy": 1.0262556517496706,
|
| 486 |
+
"epoch": 0.15578923941467757,
|
| 487 |
+
"grad_norm": 41.0,
|
| 488 |
+
"learning_rate": 6.98e-06,
|
| 489 |
+
"loss": 16.1234,
|
| 490 |
+
"mean_token_accuracy": 0.7529099363833666,
|
| 491 |
+
"num_tokens": 2320358.0,
|
| 492 |
+
"step": 350
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"entropy": 0.9883157461881638,
|
| 496 |
+
"epoch": 0.16024036054081123,
|
| 497 |
+
"grad_norm": 44.25,
|
| 498 |
+
"learning_rate": 7.180000000000001e-06,
|
| 499 |
+
"loss": 16.301,
|
| 500 |
+
"mean_token_accuracy": 0.7517775129526854,
|
| 501 |
+
"num_tokens": 2388824.0,
|
| 502 |
+
"step": 360
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"entropy": 0.9906315876170992,
|
| 506 |
+
"epoch": 0.16469148166694486,
|
| 507 |
+
"grad_norm": 39.75,
|
| 508 |
+
"learning_rate": 7.3800000000000005e-06,
|
| 509 |
+
"loss": 15.5904,
|
| 510 |
+
"mean_token_accuracy": 0.760966083779931,
|
| 511 |
+
"num_tokens": 2456144.0,
|
| 512 |
+
"step": 370
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"entropy": 1.0959791645407677,
|
| 516 |
+
"epoch": 0.1691426027930785,
|
| 517 |
+
"grad_norm": 37.5,
|
| 518 |
+
"learning_rate": 7.58e-06,
|
| 519 |
+
"loss": 17.7416,
|
| 520 |
+
"mean_token_accuracy": 0.7330215321853757,
|
| 521 |
+
"num_tokens": 2515325.0,
|
| 522 |
+
"step": 380
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"entropy": 1.0152791179716587,
|
| 526 |
+
"epoch": 0.17359372391921216,
|
| 527 |
+
"grad_norm": 38.25,
|
| 528 |
+
"learning_rate": 7.78e-06,
|
| 529 |
+
"loss": 16.1438,
|
| 530 |
+
"mean_token_accuracy": 0.7489995956420898,
|
| 531 |
+
"num_tokens": 2580490.0,
|
| 532 |
+
"step": 390
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"entropy": 1.015568952821195,
|
| 536 |
+
"epoch": 0.1780448450453458,
|
| 537 |
+
"grad_norm": 32.5,
|
| 538 |
+
"learning_rate": 7.980000000000002e-06,
|
| 539 |
+
"loss": 16.3986,
|
| 540 |
+
"mean_token_accuracy": 0.7477670447900892,
|
| 541 |
+
"num_tokens": 2644330.0,
|
| 542 |
+
"step": 400
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.1780448450453458,
|
| 546 |
+
"eval_biology_entropy": 1.4987669139504434,
|
| 547 |
+
"eval_biology_loss": 1.7285088300704956,
|
| 548 |
+
"eval_biology_mean_token_accuracy": 0.6236078375577927,
|
| 549 |
+
"eval_biology_num_tokens": 2644330.0,
|
| 550 |
+
"eval_biology_runtime": 48.385,
|
| 551 |
+
"eval_biology_samples_per_second": 10.334,
|
| 552 |
+
"eval_biology_steps_per_second": 10.334,
|
| 553 |
+
"step": 400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.1780448450453458,
|
| 557 |
+
"eval_chemistry_entropy": 1.0116422069072724,
|
| 558 |
+
"eval_chemistry_loss": 1.0169917345046997,
|
| 559 |
+
"eval_chemistry_mean_token_accuracy": 0.7494727990031242,
|
| 560 |
+
"eval_chemistry_num_tokens": 2644330.0,
|
| 561 |
+
"eval_chemistry_runtime": 55.9816,
|
| 562 |
+
"eval_chemistry_samples_per_second": 8.932,
|
| 563 |
+
"eval_chemistry_steps_per_second": 8.932,
|
| 564 |
+
"step": 400
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.1780448450453458,
|
| 568 |
+
"eval_math_entropy": 0.830569740831852,
|
| 569 |
+
"eval_math_loss": 0.9943051934242249,
|
| 570 |
+
"eval_math_mean_token_accuracy": 0.7620642938613892,
|
| 571 |
+
"eval_math_num_tokens": 2644330.0,
|
| 572 |
+
"eval_math_runtime": 60.3723,
|
| 573 |
+
"eval_math_samples_per_second": 8.282,
|
| 574 |
+
"eval_math_steps_per_second": 8.282,
|
| 575 |
+
"step": 400
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.1780448450453458,
|
| 579 |
+
"eval_physics_entropy": 0.9355049659013748,
|
| 580 |
+
"eval_physics_loss": 0.94325852394104,
|
| 581 |
+
"eval_physics_mean_token_accuracy": 0.7684184198379517,
|
| 582 |
+
"eval_physics_num_tokens": 2644330.0,
|
| 583 |
+
"eval_physics_runtime": 65.391,
|
| 584 |
+
"eval_physics_samples_per_second": 7.646,
|
| 585 |
+
"eval_physics_steps_per_second": 7.646,
|
| 586 |
+
"step": 400
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"entropy": 0.9359495824202895,
|
| 590 |
+
"epoch": 0.18249596617147945,
|
| 591 |
+
"grad_norm": 37.5,
|
| 592 |
+
"learning_rate": 8.18e-06,
|
| 593 |
+
"loss": 15.074,
|
| 594 |
+
"mean_token_accuracy": 0.7657283775508403,
|
| 595 |
+
"num_tokens": 2713234.0,
|
| 596 |
+
"step": 410
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"entropy": 1.0310791170224547,
|
| 600 |
+
"epoch": 0.18694708729761308,
|
| 601 |
+
"grad_norm": 36.75,
|
| 602 |
+
"learning_rate": 8.380000000000001e-06,
|
| 603 |
+
"loss": 16.5812,
|
| 604 |
+
"mean_token_accuracy": 0.7489650597795844,
|
| 605 |
+
"num_tokens": 2783261.0,
|
| 606 |
+
"step": 420
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"entropy": 1.0061013143509627,
|
| 610 |
+
"epoch": 0.19139820842374672,
|
| 611 |
+
"grad_norm": 34.25,
|
| 612 |
+
"learning_rate": 8.580000000000001e-06,
|
| 613 |
+
"loss": 16.2064,
|
| 614 |
+
"mean_token_accuracy": 0.7534871533513069,
|
| 615 |
+
"num_tokens": 2850170.0,
|
| 616 |
+
"step": 430
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"entropy": 0.9983460938557982,
|
| 620 |
+
"epoch": 0.19584932954988038,
|
| 621 |
+
"grad_norm": 36.5,
|
| 622 |
+
"learning_rate": 8.78e-06,
|
| 623 |
+
"loss": 15.9251,
|
| 624 |
+
"mean_token_accuracy": 0.7519530924037099,
|
| 625 |
+
"num_tokens": 2913700.0,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"entropy": 0.9914884481579065,
|
| 630 |
+
"epoch": 0.200300450676014,
|
| 631 |
+
"grad_norm": 43.25,
|
| 632 |
+
"learning_rate": 8.98e-06,
|
| 633 |
+
"loss": 16.0176,
|
| 634 |
+
"mean_token_accuracy": 0.7508705228567123,
|
| 635 |
+
"num_tokens": 2981212.0,
|
| 636 |
+
"step": 450
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"entropy": 0.9640501407906413,
|
| 640 |
+
"epoch": 0.20475157180214767,
|
| 641 |
+
"grad_norm": 43.75,
|
| 642 |
+
"learning_rate": 9.180000000000002e-06,
|
| 643 |
+
"loss": 15.5197,
|
| 644 |
+
"mean_token_accuracy": 0.7612664353102445,
|
| 645 |
+
"num_tokens": 3048973.0,
|
| 646 |
+
"step": 460
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"entropy": 0.9837653011083602,
|
| 650 |
+
"epoch": 0.2092026929282813,
|
| 651 |
+
"grad_norm": 32.75,
|
| 652 |
+
"learning_rate": 9.38e-06,
|
| 653 |
+
"loss": 16.0462,
|
| 654 |
+
"mean_token_accuracy": 0.7515783123672009,
|
| 655 |
+
"num_tokens": 3117033.0,
|
| 656 |
+
"step": 470
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"entropy": 0.9881603302434087,
|
| 660 |
+
"epoch": 0.21365381405441494,
|
| 661 |
+
"grad_norm": 35.5,
|
| 662 |
+
"learning_rate": 9.58e-06,
|
| 663 |
+
"loss": 15.6722,
|
| 664 |
+
"mean_token_accuracy": 0.7575565822422504,
|
| 665 |
+
"num_tokens": 3185255.0,
|
| 666 |
+
"step": 480
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"entropy": 0.960085684619844,
|
| 670 |
+
"epoch": 0.2181049351805486,
|
| 671 |
+
"grad_norm": 43.5,
|
| 672 |
+
"learning_rate": 9.780000000000001e-06,
|
| 673 |
+
"loss": 15.732,
|
| 674 |
+
"mean_token_accuracy": 0.7576891854405403,
|
| 675 |
+
"num_tokens": 3253121.0,
|
| 676 |
+
"step": 490
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"entropy": 0.9170863745734096,
|
| 680 |
+
"epoch": 0.22255605630668224,
|
| 681 |
+
"grad_norm": 39.5,
|
| 682 |
+
"learning_rate": 9.980000000000001e-06,
|
| 683 |
+
"loss": 14.4657,
|
| 684 |
+
"mean_token_accuracy": 0.7695914417505264,
|
| 685 |
+
"num_tokens": 3322823.0,
|
| 686 |
+
"step": 500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.22255605630668224,
|
| 690 |
+
"eval_biology_entropy": 1.4002713676095009,
|
| 691 |
+
"eval_biology_loss": 1.6903640031814575,
|
| 692 |
+
"eval_biology_mean_token_accuracy": 0.6292068202495575,
|
| 693 |
+
"eval_biology_num_tokens": 3322823.0,
|
| 694 |
+
"eval_biology_runtime": 48.4234,
|
| 695 |
+
"eval_biology_samples_per_second": 10.326,
|
| 696 |
+
"eval_biology_steps_per_second": 10.326,
|
| 697 |
+
"step": 500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.22255605630668224,
|
| 701 |
+
"eval_chemistry_entropy": 0.9267486627101899,
|
| 702 |
+
"eval_chemistry_loss": 0.9770777821540833,
|
| 703 |
+
"eval_chemistry_mean_token_accuracy": 0.7565804206728935,
|
| 704 |
+
"eval_chemistry_num_tokens": 3322823.0,
|
| 705 |
+
"eval_chemistry_runtime": 56.3359,
|
| 706 |
+
"eval_chemistry_samples_per_second": 8.875,
|
| 707 |
+
"eval_chemistry_steps_per_second": 8.875,
|
| 708 |
+
"step": 500
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.22255605630668224,
|
| 712 |
+
"eval_math_entropy": 0.7625217022895813,
|
| 713 |
+
"eval_math_loss": 0.9808707237243652,
|
| 714 |
+
"eval_math_mean_token_accuracy": 0.7647298012971878,
|
| 715 |
+
"eval_math_num_tokens": 3322823.0,
|
| 716 |
+
"eval_math_runtime": 57.8095,
|
| 717 |
+
"eval_math_samples_per_second": 8.649,
|
| 718 |
+
"eval_math_steps_per_second": 8.649,
|
| 719 |
+
"step": 500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.22255605630668224,
|
| 723 |
+
"eval_physics_entropy": 0.8588201667070389,
|
| 724 |
+
"eval_physics_loss": 0.910319983959198,
|
| 725 |
+
"eval_physics_mean_token_accuracy": 0.7743916043043136,
|
| 726 |
+
"eval_physics_num_tokens": 3322823.0,
|
| 727 |
+
"eval_physics_runtime": 64.6422,
|
| 728 |
+
"eval_physics_samples_per_second": 7.735,
|
| 729 |
+
"eval_physics_steps_per_second": 7.735,
|
| 730 |
+
"step": 500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 0.9297409053891897,
|
| 734 |
+
"epoch": 0.2270071774328159,
|
| 735 |
+
"grad_norm": 34.0,
|
| 736 |
+
"learning_rate": 1.018e-05,
|
| 737 |
+
"loss": 15.1678,
|
| 738 |
+
"mean_token_accuracy": 0.7620427209883929,
|
| 739 |
+
"num_tokens": 3393393.0,
|
| 740 |
+
"step": 510
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 1.017524036206305,
|
| 744 |
+
"epoch": 0.23145829855894953,
|
| 745 |
+
"grad_norm": 41.75,
|
| 746 |
+
"learning_rate": 1.038e-05,
|
| 747 |
+
"loss": 16.5918,
|
| 748 |
+
"mean_token_accuracy": 0.7439933486282826,
|
| 749 |
+
"num_tokens": 3454750.0,
|
| 750 |
+
"step": 520
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 0.9231159623712302,
|
| 754 |
+
"epoch": 0.2359094196850832,
|
| 755 |
+
"grad_norm": 43.5,
|
| 756 |
+
"learning_rate": 1.0580000000000002e-05,
|
| 757 |
+
"loss": 14.7926,
|
| 758 |
+
"mean_token_accuracy": 0.7705453760921955,
|
| 759 |
+
"num_tokens": 3526914.0,
|
| 760 |
+
"step": 530
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 1.0112015223130584,
|
| 764 |
+
"epoch": 0.24036054081121683,
|
| 765 |
+
"grad_norm": 42.25,
|
| 766 |
+
"learning_rate": 1.0780000000000002e-05,
|
| 767 |
+
"loss": 16.3174,
|
| 768 |
+
"mean_token_accuracy": 0.7517922900617122,
|
| 769 |
+
"num_tokens": 3589449.0,
|
| 770 |
+
"step": 540
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 0.9775515835732221,
|
| 774 |
+
"epoch": 0.24481166193735046,
|
| 775 |
+
"grad_norm": 37.5,
|
| 776 |
+
"learning_rate": 1.0980000000000002e-05,
|
| 777 |
+
"loss": 15.5292,
|
| 778 |
+
"mean_token_accuracy": 0.7556810062378645,
|
| 779 |
+
"num_tokens": 3655092.0,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 0.9711876364424825,
|
| 784 |
+
"epoch": 0.24926278306348412,
|
| 785 |
+
"grad_norm": 42.75,
|
| 786 |
+
"learning_rate": 1.1180000000000001e-05,
|
| 787 |
+
"loss": 15.8816,
|
| 788 |
+
"mean_token_accuracy": 0.7561034083366394,
|
| 789 |
+
"num_tokens": 3719113.0,
|
| 790 |
+
"step": 560
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 0.9615386385470629,
|
| 794 |
+
"epoch": 0.25371390418961776,
|
| 795 |
+
"grad_norm": 33.75,
|
| 796 |
+
"learning_rate": 1.138e-05,
|
| 797 |
+
"loss": 15.6077,
|
| 798 |
+
"mean_token_accuracy": 0.7553648345172406,
|
| 799 |
+
"num_tokens": 3785282.0,
|
| 800 |
+
"step": 570
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.9940539756789804,
|
| 804 |
+
"epoch": 0.2581650253157514,
|
| 805 |
+
"grad_norm": 31.5,
|
| 806 |
+
"learning_rate": 1.1580000000000001e-05,
|
| 807 |
+
"loss": 15.7863,
|
| 808 |
+
"mean_token_accuracy": 0.758138832077384,
|
| 809 |
+
"num_tokens": 3852276.0,
|
| 810 |
+
"step": 580
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 0.8664842249825597,
|
| 814 |
+
"epoch": 0.262616146441885,
|
| 815 |
+
"grad_norm": 32.75,
|
| 816 |
+
"learning_rate": 1.178e-05,
|
| 817 |
+
"loss": 14.4448,
|
| 818 |
+
"mean_token_accuracy": 0.7707123512402176,
|
| 819 |
+
"num_tokens": 3925649.0,
|
| 820 |
+
"step": 590
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 0.9649652477353812,
|
| 824 |
+
"epoch": 0.2670672675680187,
|
| 825 |
+
"grad_norm": 39.25,
|
| 826 |
+
"learning_rate": 1.198e-05,
|
| 827 |
+
"loss": 15.1734,
|
| 828 |
+
"mean_token_accuracy": 0.7605622936040163,
|
| 829 |
+
"num_tokens": 3990505.0,
|
| 830 |
+
"step": 600
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.2670672675680187,
|
| 834 |
+
"eval_biology_entropy": 1.3548224151134491,
|
| 835 |
+
"eval_biology_loss": 1.6628271341323853,
|
| 836 |
+
"eval_biology_mean_token_accuracy": 0.6329956727027893,
|
| 837 |
+
"eval_biology_num_tokens": 3990505.0,
|
| 838 |
+
"eval_biology_runtime": 48.5951,
|
| 839 |
+
"eval_biology_samples_per_second": 10.289,
|
| 840 |
+
"eval_biology_steps_per_second": 10.289,
|
| 841 |
+
"step": 600
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.2670672675680187,
|
| 845 |
+
"eval_chemistry_entropy": 0.8806400309503079,
|
| 846 |
+
"eval_chemistry_loss": 0.9497475028038025,
|
| 847 |
+
"eval_chemistry_mean_token_accuracy": 0.7612205193042755,
|
| 848 |
+
"eval_chemistry_num_tokens": 3990505.0,
|
| 849 |
+
"eval_chemistry_runtime": 56.0415,
|
| 850 |
+
"eval_chemistry_samples_per_second": 8.922,
|
| 851 |
+
"eval_chemistry_steps_per_second": 8.922,
|
| 852 |
+
"step": 600
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.2670672675680187,
|
| 856 |
+
"eval_math_entropy": 0.7478249241113663,
|
| 857 |
+
"eval_math_loss": 0.9696071743965149,
|
| 858 |
+
"eval_math_mean_token_accuracy": 0.7666747701764107,
|
| 859 |
+
"eval_math_num_tokens": 3990505.0,
|
| 860 |
+
"eval_math_runtime": 57.7058,
|
| 861 |
+
"eval_math_samples_per_second": 8.665,
|
| 862 |
+
"eval_math_steps_per_second": 8.665,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.2670672675680187,
|
| 867 |
+
"eval_physics_entropy": 0.8181254861354827,
|
| 868 |
+
"eval_physics_loss": 0.8910531997680664,
|
| 869 |
+
"eval_physics_mean_token_accuracy": 0.7779091320037842,
|
| 870 |
+
"eval_physics_num_tokens": 3990505.0,
|
| 871 |
+
"eval_physics_runtime": 64.595,
|
| 872 |
+
"eval_physics_samples_per_second": 7.741,
|
| 873 |
+
"eval_physics_steps_per_second": 7.741,
|
| 874 |
+
"step": 600
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"entropy": 0.9066566737368703,
|
| 878 |
+
"epoch": 0.27151838869415235,
|
| 879 |
+
"grad_norm": 35.5,
|
| 880 |
+
"learning_rate": 1.218e-05,
|
| 881 |
+
"loss": 14.8095,
|
| 882 |
+
"mean_token_accuracy": 0.7637650985270739,
|
| 883 |
+
"num_tokens": 4060657.0,
|
| 884 |
+
"step": 610
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"entropy": 0.8951275391504169,
|
| 888 |
+
"epoch": 0.275969509820286,
|
| 889 |
+
"grad_norm": 31.625,
|
| 890 |
+
"learning_rate": 1.2380000000000002e-05,
|
| 891 |
+
"loss": 14.3937,
|
| 892 |
+
"mean_token_accuracy": 0.7713713925331831,
|
| 893 |
+
"num_tokens": 4133119.0,
|
| 894 |
+
"step": 620
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"entropy": 0.9309091791510582,
|
| 898 |
+
"epoch": 0.2804206309464196,
|
| 899 |
+
"grad_norm": 40.25,
|
| 900 |
+
"learning_rate": 1.2580000000000002e-05,
|
| 901 |
+
"loss": 14.9362,
|
| 902 |
+
"mean_token_accuracy": 0.7651385851204395,
|
| 903 |
+
"num_tokens": 4200051.0,
|
| 904 |
+
"step": 630
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"entropy": 0.9745316403917968,
|
| 908 |
+
"epoch": 0.2848717520725533,
|
| 909 |
+
"grad_norm": 41.0,
|
| 910 |
+
"learning_rate": 1.2780000000000001e-05,
|
| 911 |
+
"loss": 15.86,
|
| 912 |
+
"mean_token_accuracy": 0.7543578896671533,
|
| 913 |
+
"num_tokens": 4267403.0,
|
| 914 |
+
"step": 640
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"entropy": 0.9337521677836775,
|
| 918 |
+
"epoch": 0.28932287319868694,
|
| 919 |
+
"grad_norm": 39.0,
|
| 920 |
+
"learning_rate": 1.2980000000000001e-05,
|
| 921 |
+
"loss": 14.9738,
|
| 922 |
+
"mean_token_accuracy": 0.7643170915544033,
|
| 923 |
+
"num_tokens": 4333268.0,
|
| 924 |
+
"step": 650
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"entropy": 0.9018357262946666,
|
| 928 |
+
"epoch": 0.29377399432482054,
|
| 929 |
+
"grad_norm": 53.5,
|
| 930 |
+
"learning_rate": 1.3180000000000001e-05,
|
| 931 |
+
"loss": 14.5097,
|
| 932 |
+
"mean_token_accuracy": 0.7716417364776135,
|
| 933 |
+
"num_tokens": 4400242.0,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"entropy": 0.9362477554008365,
|
| 938 |
+
"epoch": 0.2982251154509542,
|
| 939 |
+
"grad_norm": 34.5,
|
| 940 |
+
"learning_rate": 1.3380000000000002e-05,
|
| 941 |
+
"loss": 15.1358,
|
| 942 |
+
"mean_token_accuracy": 0.7618589885532856,
|
| 943 |
+
"num_tokens": 4464714.0,
|
| 944 |
+
"step": 670
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"entropy": 0.8689518304541707,
|
| 948 |
+
"epoch": 0.30267623657708786,
|
| 949 |
+
"grad_norm": 41.25,
|
| 950 |
+
"learning_rate": 1.3580000000000002e-05,
|
| 951 |
+
"loss": 14.3018,
|
| 952 |
+
"mean_token_accuracy": 0.7737269978970289,
|
| 953 |
+
"num_tokens": 4535458.0,
|
| 954 |
+
"step": 680
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"entropy": 0.9227319629862905,
|
| 958 |
+
"epoch": 0.3071273577032215,
|
| 959 |
+
"grad_norm": 38.25,
|
| 960 |
+
"learning_rate": 1.378e-05,
|
| 961 |
+
"loss": 14.6416,
|
| 962 |
+
"mean_token_accuracy": 0.7681682342663407,
|
| 963 |
+
"num_tokens": 4599099.0,
|
| 964 |
+
"step": 690
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"entropy": 0.865695190615952,
|
| 968 |
+
"epoch": 0.31157847882935513,
|
| 969 |
+
"grad_norm": 28.25,
|
| 970 |
+
"learning_rate": 1.398e-05,
|
| 971 |
+
"loss": 14.0258,
|
| 972 |
+
"mean_token_accuracy": 0.7762727830559015,
|
| 973 |
+
"num_tokens": 4662903.0,
|
| 974 |
+
"step": 700
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.31157847882935513,
|
| 978 |
+
"eval_biology_entropy": 1.4728712896108627,
|
| 979 |
+
"eval_biology_loss": 1.6218550205230713,
|
| 980 |
+
"eval_biology_mean_token_accuracy": 0.6382006840705872,
|
| 981 |
+
"eval_biology_num_tokens": 4662903.0,
|
| 982 |
+
"eval_biology_runtime": 48.5605,
|
| 983 |
+
"eval_biology_samples_per_second": 10.296,
|
| 984 |
+
"eval_biology_steps_per_second": 10.296,
|
| 985 |
+
"step": 700
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.31157847882935513,
|
| 989 |
+
"eval_chemistry_entropy": 0.9336612664461136,
|
| 990 |
+
"eval_chemistry_loss": 0.9250124096870422,
|
| 991 |
+
"eval_chemistry_mean_token_accuracy": 0.7656691336631775,
|
| 992 |
+
"eval_chemistry_num_tokens": 4662903.0,
|
| 993 |
+
"eval_chemistry_runtime": 56.0811,
|
| 994 |
+
"eval_chemistry_samples_per_second": 8.916,
|
| 995 |
+
"eval_chemistry_steps_per_second": 8.916,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.31157847882935513,
|
| 1000 |
+
"eval_math_entropy": 0.7846916960477829,
|
| 1001 |
+
"eval_math_loss": 0.9586912393569946,
|
| 1002 |
+
"eval_math_mean_token_accuracy": 0.7681104638576508,
|
| 1003 |
+
"eval_math_num_tokens": 4662903.0,
|
| 1004 |
+
"eval_math_runtime": 57.7641,
|
| 1005 |
+
"eval_math_samples_per_second": 8.656,
|
| 1006 |
+
"eval_math_steps_per_second": 8.656,
|
| 1007 |
+
"step": 700
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 0.31157847882935513,
|
| 1011 |
+
"eval_physics_entropy": 0.8758866502046585,
|
| 1012 |
+
"eval_physics_loss": 0.8707027435302734,
|
| 1013 |
+
"eval_physics_mean_token_accuracy": 0.7806350255012512,
|
| 1014 |
+
"eval_physics_num_tokens": 4662903.0,
|
| 1015 |
+
"eval_physics_runtime": 64.5913,
|
| 1016 |
+
"eval_physics_samples_per_second": 7.741,
|
| 1017 |
+
"eval_physics_steps_per_second": 7.741,
|
| 1018 |
+
"step": 700
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"entropy": 0.9134258929640054,
|
| 1022 |
+
"epoch": 0.3160295999554888,
|
| 1023 |
+
"grad_norm": 42.0,
|
| 1024 |
+
"learning_rate": 1.418e-05,
|
| 1025 |
+
"loss": 15.0312,
|
| 1026 |
+
"mean_token_accuracy": 0.7628084912896156,
|
| 1027 |
+
"num_tokens": 4729166.0,
|
| 1028 |
+
"step": 710
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"entropy": 0.97938490267843,
|
| 1032 |
+
"epoch": 0.32048072108162245,
|
| 1033 |
+
"grad_norm": 41.75,
|
| 1034 |
+
"learning_rate": 1.4380000000000001e-05,
|
| 1035 |
+
"loss": 15.7738,
|
| 1036 |
+
"mean_token_accuracy": 0.7518788330256939,
|
| 1037 |
+
"num_tokens": 4796815.0,
|
| 1038 |
+
"step": 720
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"entropy": 0.8446940638124942,
|
| 1042 |
+
"epoch": 0.32493184220775606,
|
| 1043 |
+
"grad_norm": 34.5,
|
| 1044 |
+
"learning_rate": 1.4580000000000001e-05,
|
| 1045 |
+
"loss": 13.5574,
|
| 1046 |
+
"mean_token_accuracy": 0.782823646441102,
|
| 1047 |
+
"num_tokens": 4865172.0,
|
| 1048 |
+
"step": 730
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"entropy": 0.9024957665242255,
|
| 1052 |
+
"epoch": 0.3293829633338897,
|
| 1053 |
+
"grad_norm": 33.75,
|
| 1054 |
+
"learning_rate": 1.478e-05,
|
| 1055 |
+
"loss": 14.8961,
|
| 1056 |
+
"mean_token_accuracy": 0.763876224681735,
|
| 1057 |
+
"num_tokens": 4932671.0,
|
| 1058 |
+
"step": 740
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"entropy": 0.9568601671606303,
|
| 1062 |
+
"epoch": 0.3338340844600234,
|
| 1063 |
+
"grad_norm": 33.25,
|
| 1064 |
+
"learning_rate": 1.498e-05,
|
| 1065 |
+
"loss": 15.257,
|
| 1066 |
+
"mean_token_accuracy": 0.7628579653799534,
|
| 1067 |
+
"num_tokens": 4998710.0,
|
| 1068 |
+
"step": 750
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"entropy": 0.8733517214655876,
|
| 1072 |
+
"epoch": 0.338285205586157,
|
| 1073 |
+
"grad_norm": 30.75,
|
| 1074 |
+
"learning_rate": 1.5180000000000002e-05,
|
| 1075 |
+
"loss": 14.1521,
|
| 1076 |
+
"mean_token_accuracy": 0.7736187670379877,
|
| 1077 |
+
"num_tokens": 5066948.0,
|
| 1078 |
+
"step": 760
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"entropy": 0.9070882228203118,
|
| 1082 |
+
"epoch": 0.34273632671229065,
|
| 1083 |
+
"grad_norm": 35.25,
|
| 1084 |
+
"learning_rate": 1.5380000000000002e-05,
|
| 1085 |
+
"loss": 14.7033,
|
| 1086 |
+
"mean_token_accuracy": 0.7682277161628008,
|
| 1087 |
+
"num_tokens": 5129950.0,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"entropy": 0.9775535127148032,
|
| 1092 |
+
"epoch": 0.3471874478384243,
|
| 1093 |
+
"grad_norm": 34.5,
|
| 1094 |
+
"learning_rate": 1.5580000000000003e-05,
|
| 1095 |
+
"loss": 15.453,
|
| 1096 |
+
"mean_token_accuracy": 0.754616218432784,
|
| 1097 |
+
"num_tokens": 5193519.0,
|
| 1098 |
+
"step": 780
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"entropy": 0.8829231994226575,
|
| 1102 |
+
"epoch": 0.351638568964558,
|
| 1103 |
+
"grad_norm": 41.75,
|
| 1104 |
+
"learning_rate": 1.578e-05,
|
| 1105 |
+
"loss": 14.6374,
|
| 1106 |
+
"mean_token_accuracy": 0.7640287645161152,
|
| 1107 |
+
"num_tokens": 5256503.0,
|
| 1108 |
+
"step": 790
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"entropy": 0.9582383582368493,
|
| 1112 |
+
"epoch": 0.3560896900906916,
|
| 1113 |
+
"grad_norm": 40.25,
|
| 1114 |
+
"learning_rate": 1.5980000000000003e-05,
|
| 1115 |
+
"loss": 15.6331,
|
| 1116 |
+
"mean_token_accuracy": 0.7566823102533817,
|
| 1117 |
+
"num_tokens": 5324751.0,
|
| 1118 |
+
"step": 800
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.3560896900906916,
|
| 1122 |
+
"eval_biology_entropy": 1.554333319604397,
|
| 1123 |
+
"eval_biology_loss": 1.5954139232635498,
|
| 1124 |
+
"eval_biology_mean_token_accuracy": 0.6430081333518028,
|
| 1125 |
+
"eval_biology_num_tokens": 5324751.0,
|
| 1126 |
+
"eval_biology_runtime": 49.1621,
|
| 1127 |
+
"eval_biology_samples_per_second": 10.17,
|
| 1128 |
+
"eval_biology_steps_per_second": 10.17,
|
| 1129 |
+
"step": 800
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.3560896900906916,
|
| 1133 |
+
"eval_chemistry_entropy": 0.9556231063604355,
|
| 1134 |
+
"eval_chemistry_loss": 0.907728910446167,
|
| 1135 |
+
"eval_chemistry_mean_token_accuracy": 0.7686411096453667,
|
| 1136 |
+
"eval_chemistry_num_tokens": 5324751.0,
|
| 1137 |
+
"eval_chemistry_runtime": 56.1564,
|
| 1138 |
+
"eval_chemistry_samples_per_second": 8.904,
|
| 1139 |
+
"eval_chemistry_steps_per_second": 8.904,
|
| 1140 |
+
"step": 800
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 0.3560896900906916,
|
| 1144 |
+
"eval_math_entropy": 0.7810768990516662,
|
| 1145 |
+
"eval_math_loss": 0.949228048324585,
|
| 1146 |
+
"eval_math_mean_token_accuracy": 0.7701822088956833,
|
| 1147 |
+
"eval_math_num_tokens": 5324751.0,
|
| 1148 |
+
"eval_math_runtime": 57.7577,
|
| 1149 |
+
"eval_math_samples_per_second": 8.657,
|
| 1150 |
+
"eval_math_steps_per_second": 8.657,
|
| 1151 |
+
"step": 800
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.3560896900906916,
|
| 1155 |
+
"eval_physics_entropy": 0.895748406291008,
|
| 1156 |
+
"eval_physics_loss": 0.8605655431747437,
|
| 1157 |
+
"eval_physics_mean_token_accuracy": 0.7827676599621772,
|
| 1158 |
+
"eval_physics_num_tokens": 5324751.0,
|
| 1159 |
+
"eval_physics_runtime": 64.6674,
|
| 1160 |
+
"eval_physics_samples_per_second": 7.732,
|
| 1161 |
+
"eval_physics_steps_per_second": 7.732,
|
| 1162 |
+
"step": 800
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"entropy": 0.8505169737152756,
|
| 1166 |
+
"epoch": 0.36054081121682524,
|
| 1167 |
+
"grad_norm": 41.0,
|
| 1168 |
+
"learning_rate": 1.618e-05,
|
| 1169 |
+
"loss": 13.7198,
|
| 1170 |
+
"mean_token_accuracy": 0.7793542932718992,
|
| 1171 |
+
"num_tokens": 5391664.0,
|
| 1172 |
+
"step": 810
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"entropy": 0.8401029847562314,
|
| 1176 |
+
"epoch": 0.3649919323429589,
|
| 1177 |
+
"grad_norm": 33.25,
|
| 1178 |
+
"learning_rate": 1.638e-05,
|
| 1179 |
+
"loss": 13.3589,
|
| 1180 |
+
"mean_token_accuracy": 0.7823891170322895,
|
| 1181 |
+
"num_tokens": 5458519.0,
|
| 1182 |
+
"step": 820
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"entropy": 0.8939393721520901,
|
| 1186 |
+
"epoch": 0.3694430534690925,
|
| 1187 |
+
"grad_norm": 36.25,
|
| 1188 |
+
"learning_rate": 1.658e-05,
|
| 1189 |
+
"loss": 14.7591,
|
| 1190 |
+
"mean_token_accuracy": 0.7666325703263283,
|
| 1191 |
+
"num_tokens": 5523391.0,
|
| 1192 |
+
"step": 830
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"entropy": 0.8917801808565855,
|
| 1196 |
+
"epoch": 0.37389417459522617,
|
| 1197 |
+
"grad_norm": 30.375,
|
| 1198 |
+
"learning_rate": 1.6780000000000002e-05,
|
| 1199 |
+
"loss": 14.1075,
|
| 1200 |
+
"mean_token_accuracy": 0.7729504836723209,
|
| 1201 |
+
"num_tokens": 5585508.0,
|
| 1202 |
+
"step": 840
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"entropy": 0.8576251868158579,
|
| 1206 |
+
"epoch": 0.37834529572135983,
|
| 1207 |
+
"grad_norm": 38.25,
|
| 1208 |
+
"learning_rate": 1.698e-05,
|
| 1209 |
+
"loss": 14.3124,
|
| 1210 |
+
"mean_token_accuracy": 0.774137607589364,
|
| 1211 |
+
"num_tokens": 5650755.0,
|
| 1212 |
+
"step": 850
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"entropy": 0.8919966861605644,
|
| 1216 |
+
"epoch": 0.38279641684749344,
|
| 1217 |
+
"grad_norm": 33.75,
|
| 1218 |
+
"learning_rate": 1.718e-05,
|
| 1219 |
+
"loss": 14.1824,
|
| 1220 |
+
"mean_token_accuracy": 0.7732574883848429,
|
| 1221 |
+
"num_tokens": 5716321.0,
|
| 1222 |
+
"step": 860
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"entropy": 0.9272915873676538,
|
| 1226 |
+
"epoch": 0.3872475379736271,
|
| 1227 |
+
"grad_norm": 42.25,
|
| 1228 |
+
"learning_rate": 1.7380000000000003e-05,
|
| 1229 |
+
"loss": 14.8897,
|
| 1230 |
+
"mean_token_accuracy": 0.7641183525323868,
|
| 1231 |
+
"num_tokens": 5782698.0,
|
| 1232 |
+
"step": 870
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"entropy": 0.8425593223422766,
|
| 1236 |
+
"epoch": 0.39169865909976076,
|
| 1237 |
+
"grad_norm": 30.0,
|
| 1238 |
+
"learning_rate": 1.758e-05,
|
| 1239 |
+
"loss": 13.8179,
|
| 1240 |
+
"mean_token_accuracy": 0.7790724016726017,
|
| 1241 |
+
"num_tokens": 5848889.0,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"entropy": 0.8974415950477124,
|
| 1246 |
+
"epoch": 0.3961497802258944,
|
| 1247 |
+
"grad_norm": 34.25,
|
| 1248 |
+
"learning_rate": 1.7780000000000003e-05,
|
| 1249 |
+
"loss": 14.5537,
|
| 1250 |
+
"mean_token_accuracy": 0.7696893479675054,
|
| 1251 |
+
"num_tokens": 5912919.0,
|
| 1252 |
+
"step": 890
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"entropy": 0.868491224758327,
|
| 1256 |
+
"epoch": 0.400600901352028,
|
| 1257 |
+
"grad_norm": 29.5,
|
| 1258 |
+
"learning_rate": 1.798e-05,
|
| 1259 |
+
"loss": 14.1228,
|
| 1260 |
+
"mean_token_accuracy": 0.7739411123096943,
|
| 1261 |
+
"num_tokens": 5980229.0,
|
| 1262 |
+
"step": 900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.400600901352028,
|
| 1266 |
+
"eval_biology_entropy": 1.4719535266160966,
|
| 1267 |
+
"eval_biology_loss": 1.5840708017349243,
|
| 1268 |
+
"eval_biology_mean_token_accuracy": 0.6444517712593079,
|
| 1269 |
+
"eval_biology_num_tokens": 5980229.0,
|
| 1270 |
+
"eval_biology_runtime": 48.3845,
|
| 1271 |
+
"eval_biology_samples_per_second": 10.334,
|
| 1272 |
+
"eval_biology_steps_per_second": 10.334,
|
| 1273 |
+
"step": 900
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.400600901352028,
|
| 1277 |
+
"eval_chemistry_entropy": 0.9006564086675644,
|
| 1278 |
+
"eval_chemistry_loss": 0.8931591510772705,
|
| 1279 |
+
"eval_chemistry_mean_token_accuracy": 0.771476098716259,
|
| 1280 |
+
"eval_chemistry_num_tokens": 5980229.0,
|
| 1281 |
+
"eval_chemistry_runtime": 55.9522,
|
| 1282 |
+
"eval_chemistry_samples_per_second": 8.936,
|
| 1283 |
+
"eval_chemistry_steps_per_second": 8.936,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.400600901352028,
|
| 1288 |
+
"eval_math_entropy": 0.7833816863894463,
|
| 1289 |
+
"eval_math_loss": 0.9406895637512207,
|
| 1290 |
+
"eval_math_mean_token_accuracy": 0.7704955268502235,
|
| 1291 |
+
"eval_math_num_tokens": 5980229.0,
|
| 1292 |
+
"eval_math_runtime": 57.6393,
|
| 1293 |
+
"eval_math_samples_per_second": 8.675,
|
| 1294 |
+
"eval_math_steps_per_second": 8.675,
|
| 1295 |
+
"step": 900
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 0.400600901352028,
|
| 1299 |
+
"eval_physics_entropy": 0.8495812131166458,
|
| 1300 |
+
"eval_physics_loss": 0.8490710258483887,
|
| 1301 |
+
"eval_physics_mean_token_accuracy": 0.7849877900481224,
|
| 1302 |
+
"eval_physics_num_tokens": 5980229.0,
|
| 1303 |
+
"eval_physics_runtime": 64.5379,
|
| 1304 |
+
"eval_physics_samples_per_second": 7.747,
|
| 1305 |
+
"eval_physics_steps_per_second": 7.747,
|
| 1306 |
+
"step": 900
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"entropy": 0.8243395276367664,
|
| 1310 |
+
"epoch": 0.4050520224781617,
|
| 1311 |
+
"grad_norm": 30.5,
|
| 1312 |
+
"learning_rate": 1.8180000000000002e-05,
|
| 1313 |
+
"loss": 13.4007,
|
| 1314 |
+
"mean_token_accuracy": 0.7826393615454436,
|
| 1315 |
+
"num_tokens": 6048508.0,
|
| 1316 |
+
"step": 910
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"entropy": 0.8578345064073801,
|
| 1320 |
+
"epoch": 0.40950314360429535,
|
| 1321 |
+
"grad_norm": 32.5,
|
| 1322 |
+
"learning_rate": 1.8380000000000004e-05,
|
| 1323 |
+
"loss": 13.8874,
|
| 1324 |
+
"mean_token_accuracy": 0.7774036221206189,
|
| 1325 |
+
"num_tokens": 6114855.0,
|
| 1326 |
+
"step": 920
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"entropy": 0.8919947864487767,
|
| 1330 |
+
"epoch": 0.41395426473042896,
|
| 1331 |
+
"grad_norm": 33.5,
|
| 1332 |
+
"learning_rate": 1.858e-05,
|
| 1333 |
+
"loss": 14.3682,
|
| 1334 |
+
"mean_token_accuracy": 0.773763508349657,
|
| 1335 |
+
"num_tokens": 6179050.0,
|
| 1336 |
+
"step": 930
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"entropy": 0.877837923169136,
|
| 1340 |
+
"epoch": 0.4184053858565626,
|
| 1341 |
+
"grad_norm": 33.5,
|
| 1342 |
+
"learning_rate": 1.878e-05,
|
| 1343 |
+
"loss": 14.1955,
|
| 1344 |
+
"mean_token_accuracy": 0.7732131693512201,
|
| 1345 |
+
"num_tokens": 6244987.0,
|
| 1346 |
+
"step": 940
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"entropy": 0.8329715754836797,
|
| 1350 |
+
"epoch": 0.4228565069826963,
|
| 1351 |
+
"grad_norm": 41.0,
|
| 1352 |
+
"learning_rate": 1.898e-05,
|
| 1353 |
+
"loss": 13.5939,
|
| 1354 |
+
"mean_token_accuracy": 0.7785252593457699,
|
| 1355 |
+
"num_tokens": 6310799.0,
|
| 1356 |
+
"step": 950
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"entropy": 0.8469648731872439,
|
| 1360 |
+
"epoch": 0.4273076281088299,
|
| 1361 |
+
"grad_norm": 36.75,
|
| 1362 |
+
"learning_rate": 1.918e-05,
|
| 1363 |
+
"loss": 13.6875,
|
| 1364 |
+
"mean_token_accuracy": 0.7780090972781182,
|
| 1365 |
+
"num_tokens": 6378152.0,
|
| 1366 |
+
"step": 960
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"entropy": 0.8630953669548035,
|
| 1370 |
+
"epoch": 0.43175874923496355,
|
| 1371 |
+
"grad_norm": 34.0,
|
| 1372 |
+
"learning_rate": 1.938e-05,
|
| 1373 |
+
"loss": 13.8631,
|
| 1374 |
+
"mean_token_accuracy": 0.7757172901183367,
|
| 1375 |
+
"num_tokens": 6442491.0,
|
| 1376 |
+
"step": 970
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"entropy": 0.907545124553144,
|
| 1380 |
+
"epoch": 0.4362098703610972,
|
| 1381 |
+
"grad_norm": 28.75,
|
| 1382 |
+
"learning_rate": 1.9580000000000002e-05,
|
| 1383 |
+
"loss": 14.7101,
|
| 1384 |
+
"mean_token_accuracy": 0.766944982856512,
|
| 1385 |
+
"num_tokens": 6510231.0,
|
| 1386 |
+
"step": 980
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"entropy": 0.8721130631864071,
|
| 1390 |
+
"epoch": 0.44066099148723087,
|
| 1391 |
+
"grad_norm": 34.25,
|
| 1392 |
+
"learning_rate": 1.978e-05,
|
| 1393 |
+
"loss": 14.193,
|
| 1394 |
+
"mean_token_accuracy": 0.7728719780221581,
|
| 1395 |
+
"num_tokens": 6572858.0,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"entropy": 0.8767424447461962,
|
| 1400 |
+
"epoch": 0.4451121126133645,
|
| 1401 |
+
"grad_norm": 24.625,
|
| 1402 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 1403 |
+
"loss": 14.338,
|
| 1404 |
+
"mean_token_accuracy": 0.7690305080264807,
|
| 1405 |
+
"num_tokens": 6637273.0,
|
| 1406 |
+
"step": 1000
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 0.4451121126133645,
|
| 1410 |
+
"eval_biology_entropy": 1.4690953947901726,
|
| 1411 |
+
"eval_biology_loss": 1.5712339878082275,
|
| 1412 |
+
"eval_biology_mean_token_accuracy": 0.6474293968081475,
|
| 1413 |
+
"eval_biology_num_tokens": 6637273.0,
|
| 1414 |
+
"eval_biology_runtime": 48.4474,
|
| 1415 |
+
"eval_biology_samples_per_second": 10.32,
|
| 1416 |
+
"eval_biology_steps_per_second": 10.32,
|
| 1417 |
+
"step": 1000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.4451121126133645,
|
| 1421 |
+
"eval_chemistry_entropy": 0.89704896864295,
|
| 1422 |
+
"eval_chemistry_loss": 0.8829187154769897,
|
| 1423 |
+
"eval_chemistry_mean_token_accuracy": 0.7738431564569473,
|
| 1424 |
+
"eval_chemistry_num_tokens": 6637273.0,
|
| 1425 |
+
"eval_chemistry_runtime": 56.1772,
|
| 1426 |
+
"eval_chemistry_samples_per_second": 8.9,
|
| 1427 |
+
"eval_chemistry_steps_per_second": 8.9,
|
| 1428 |
+
"step": 1000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 0.4451121126133645,
|
| 1432 |
+
"eval_math_entropy": 0.7748903884291649,
|
| 1433 |
+
"eval_math_loss": 0.9362199902534485,
|
| 1434 |
+
"eval_math_mean_token_accuracy": 0.7724022324085236,
|
| 1435 |
+
"eval_math_num_tokens": 6637273.0,
|
| 1436 |
+
"eval_math_runtime": 57.7189,
|
| 1437 |
+
"eval_math_samples_per_second": 8.663,
|
| 1438 |
+
"eval_math_steps_per_second": 8.663,
|
| 1439 |
+
"step": 1000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.4451121126133645,
|
| 1443 |
+
"eval_physics_entropy": 0.8675668377280236,
|
| 1444 |
+
"eval_physics_loss": 0.8455091118812561,
|
| 1445 |
+
"eval_physics_mean_token_accuracy": 0.7849516254663468,
|
| 1446 |
+
"eval_physics_num_tokens": 6637273.0,
|
| 1447 |
+
"eval_physics_runtime": 64.6865,
|
| 1448 |
+
"eval_physics_samples_per_second": 7.73,
|
| 1449 |
+
"eval_physics_steps_per_second": 7.73,
|
| 1450 |
+
"step": 1000
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"entropy": 0.8708374679088593,
|
| 1454 |
+
"epoch": 0.44956323373949814,
|
| 1455 |
+
"grad_norm": 33.25,
|
| 1456 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 1457 |
+
"loss": 13.9302,
|
| 1458 |
+
"mean_token_accuracy": 0.774122205376625,
|
| 1459 |
+
"num_tokens": 6700799.0,
|
| 1460 |
+
"step": 1010
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"entropy": 0.8613243264146149,
|
| 1464 |
+
"epoch": 0.4540143548656318,
|
| 1465 |
+
"grad_norm": 32.0,
|
| 1466 |
+
"learning_rate": 1.995777777777778e-05,
|
| 1467 |
+
"loss": 14.124,
|
| 1468 |
+
"mean_token_accuracy": 0.776879669725895,
|
| 1469 |
+
"num_tokens": 6765798.0,
|
| 1470 |
+
"step": 1020
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"entropy": 0.8134710913524031,
|
| 1474 |
+
"epoch": 0.4584654759917654,
|
| 1475 |
+
"grad_norm": 31.625,
|
| 1476 |
+
"learning_rate": 1.9935555555555557e-05,
|
| 1477 |
+
"loss": 13.3521,
|
| 1478 |
+
"mean_token_accuracy": 0.7832869034260511,
|
| 1479 |
+
"num_tokens": 6832774.0,
|
| 1480 |
+
"step": 1030
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"entropy": 0.890234874933958,
|
| 1484 |
+
"epoch": 0.46291659711789906,
|
| 1485 |
+
"grad_norm": 31.125,
|
| 1486 |
+
"learning_rate": 1.9913333333333335e-05,
|
| 1487 |
+
"loss": 14.2521,
|
| 1488 |
+
"mean_token_accuracy": 0.7710054367780685,
|
| 1489 |
+
"num_tokens": 6896684.0,
|
| 1490 |
+
"step": 1040
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"entropy": 0.8423492605797946,
|
| 1494 |
+
"epoch": 0.4673677182440327,
|
| 1495 |
+
"grad_norm": 40.75,
|
| 1496 |
+
"learning_rate": 1.9891111111111112e-05,
|
| 1497 |
+
"loss": 13.6826,
|
| 1498 |
+
"mean_token_accuracy": 0.7855902209877967,
|
| 1499 |
+
"num_tokens": 6964423.0,
|
| 1500 |
+
"step": 1050
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"entropy": 0.8360794227570295,
|
| 1504 |
+
"epoch": 0.4718188393701664,
|
| 1505 |
+
"grad_norm": 27.0,
|
| 1506 |
+
"learning_rate": 1.986888888888889e-05,
|
| 1507 |
+
"loss": 13.722,
|
| 1508 |
+
"mean_token_accuracy": 0.7778070000931621,
|
| 1509 |
+
"num_tokens": 7033417.0,
|
| 1510 |
+
"step": 1060
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"entropy": 0.8722811192274094,
|
| 1514 |
+
"epoch": 0.4762699604963,
|
| 1515 |
+
"grad_norm": 35.0,
|
| 1516 |
+
"learning_rate": 1.9846666666666668e-05,
|
| 1517 |
+
"loss": 14.2113,
|
| 1518 |
+
"mean_token_accuracy": 0.7714831795543432,
|
| 1519 |
+
"num_tokens": 7101274.0,
|
| 1520 |
+
"step": 1070
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"entropy": 0.8730784686282277,
|
| 1524 |
+
"epoch": 0.48072108162243365,
|
| 1525 |
+
"grad_norm": 28.5,
|
| 1526 |
+
"learning_rate": 1.9824444444444445e-05,
|
| 1527 |
+
"loss": 14.0558,
|
| 1528 |
+
"mean_token_accuracy": 0.7746166784316302,
|
| 1529 |
+
"num_tokens": 7166608.0,
|
| 1530 |
+
"step": 1080
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"entropy": 0.8250762648880482,
|
| 1534 |
+
"epoch": 0.4851722027485673,
|
| 1535 |
+
"grad_norm": 43.25,
|
| 1536 |
+
"learning_rate": 1.9802222222222226e-05,
|
| 1537 |
+
"loss": 13.4239,
|
| 1538 |
+
"mean_token_accuracy": 0.786403126642108,
|
| 1539 |
+
"num_tokens": 7240958.0,
|
| 1540 |
+
"step": 1090
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"entropy": 0.8573660463094711,
|
| 1544 |
+
"epoch": 0.4896233238747009,
|
| 1545 |
+
"grad_norm": 32.5,
|
| 1546 |
+
"learning_rate": 1.978e-05,
|
| 1547 |
+
"loss": 13.9198,
|
| 1548 |
+
"mean_token_accuracy": 0.7765211522579193,
|
| 1549 |
+
"num_tokens": 7311968.0,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.4896233238747009,
|
| 1554 |
+
"eval_biology_entropy": 1.4332191476225853,
|
| 1555 |
+
"eval_biology_loss": 1.556909203529358,
|
| 1556 |
+
"eval_biology_mean_token_accuracy": 0.6496381472945213,
|
| 1557 |
+
"eval_biology_num_tokens": 7311968.0,
|
| 1558 |
+
"eval_biology_runtime": 48.4694,
|
| 1559 |
+
"eval_biology_samples_per_second": 10.316,
|
| 1560 |
+
"eval_biology_steps_per_second": 10.316,
|
| 1561 |
+
"step": 1100
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 0.4896233238747009,
|
| 1565 |
+
"eval_chemistry_entropy": 0.8447221714258194,
|
| 1566 |
+
"eval_chemistry_loss": 0.8686630129814148,
|
| 1567 |
+
"eval_chemistry_mean_token_accuracy": 0.7762373806834221,
|
| 1568 |
+
"eval_chemistry_num_tokens": 7311968.0,
|
| 1569 |
+
"eval_chemistry_runtime": 56.0722,
|
| 1570 |
+
"eval_chemistry_samples_per_second": 8.917,
|
| 1571 |
+
"eval_chemistry_steps_per_second": 8.917,
|
| 1572 |
+
"step": 1100
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 0.4896233238747009,
|
| 1576 |
+
"eval_math_entropy": 0.7254823694527149,
|
| 1577 |
+
"eval_math_loss": 0.937345564365387,
|
| 1578 |
+
"eval_math_mean_token_accuracy": 0.7738885276317596,
|
| 1579 |
+
"eval_math_num_tokens": 7311968.0,
|
| 1580 |
+
"eval_math_runtime": 57.7068,
|
| 1581 |
+
"eval_math_samples_per_second": 8.664,
|
| 1582 |
+
"eval_math_steps_per_second": 8.664,
|
| 1583 |
+
"step": 1100
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 0.4896233238747009,
|
| 1587 |
+
"eval_physics_entropy": 0.8062724179625511,
|
| 1588 |
+
"eval_physics_loss": 0.8333404660224915,
|
| 1589 |
+
"eval_physics_mean_token_accuracy": 0.7875058565735817,
|
| 1590 |
+
"eval_physics_num_tokens": 7311968.0,
|
| 1591 |
+
"eval_physics_runtime": 65.4261,
|
| 1592 |
+
"eval_physics_samples_per_second": 7.642,
|
| 1593 |
+
"eval_physics_steps_per_second": 7.642,
|
| 1594 |
+
"step": 1100
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"entropy": 0.8604997573420405,
|
| 1598 |
+
"epoch": 0.4940744450008346,
|
| 1599 |
+
"grad_norm": 30.25,
|
| 1600 |
+
"learning_rate": 1.975777777777778e-05,
|
| 1601 |
+
"loss": 13.7608,
|
| 1602 |
+
"mean_token_accuracy": 0.7766703650355339,
|
| 1603 |
+
"num_tokens": 7378961.0,
|
| 1604 |
+
"step": 1110
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"entropy": 0.8193202435038984,
|
| 1608 |
+
"epoch": 0.49852556612696824,
|
| 1609 |
+
"grad_norm": 35.25,
|
| 1610 |
+
"learning_rate": 1.9735555555555556e-05,
|
| 1611 |
+
"loss": 13.3273,
|
| 1612 |
+
"mean_token_accuracy": 0.7826184341683984,
|
| 1613 |
+
"num_tokens": 7444923.0,
|
| 1614 |
+
"step": 1120
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"entropy": 0.8790039600804448,
|
| 1618 |
+
"epoch": 0.5029766872531019,
|
| 1619 |
+
"grad_norm": 34.0,
|
| 1620 |
+
"learning_rate": 1.9713333333333337e-05,
|
| 1621 |
+
"loss": 13.9384,
|
| 1622 |
+
"mean_token_accuracy": 0.7745813885703683,
|
| 1623 |
+
"num_tokens": 7509885.0,
|
| 1624 |
+
"step": 1130
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"entropy": 0.8574295949190855,
|
| 1628 |
+
"epoch": 0.5074278083792355,
|
| 1629 |
+
"grad_norm": 30.625,
|
| 1630 |
+
"learning_rate": 1.969111111111111e-05,
|
| 1631 |
+
"loss": 14.3156,
|
| 1632 |
+
"mean_token_accuracy": 0.7717645790427923,
|
| 1633 |
+
"num_tokens": 7576456.0,
|
| 1634 |
+
"step": 1140
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"entropy": 0.9021984472870826,
|
| 1638 |
+
"epoch": 0.5118789295053692,
|
| 1639 |
+
"grad_norm": 34.75,
|
| 1640 |
+
"learning_rate": 1.9668888888888892e-05,
|
| 1641 |
+
"loss": 14.4928,
|
| 1642 |
+
"mean_token_accuracy": 0.7711123242974282,
|
| 1643 |
+
"num_tokens": 7641821.0,
|
| 1644 |
+
"step": 1150
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"entropy": 0.8312832812778652,
|
| 1648 |
+
"epoch": 0.5163300506315028,
|
| 1649 |
+
"grad_norm": 33.5,
|
| 1650 |
+
"learning_rate": 1.9646666666666666e-05,
|
| 1651 |
+
"loss": 13.539,
|
| 1652 |
+
"mean_token_accuracy": 0.7779583852738142,
|
| 1653 |
+
"num_tokens": 7706502.0,
|
| 1654 |
+
"step": 1160
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"entropy": 0.9160518892109394,
|
| 1658 |
+
"epoch": 0.5207811717576365,
|
| 1659 |
+
"grad_norm": 35.0,
|
| 1660 |
+
"learning_rate": 1.9624444444444447e-05,
|
| 1661 |
+
"loss": 14.7744,
|
| 1662 |
+
"mean_token_accuracy": 0.7682293139398098,
|
| 1663 |
+
"num_tokens": 7772353.0,
|
| 1664 |
+
"step": 1170
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"entropy": 0.8255004599690438,
|
| 1668 |
+
"epoch": 0.52523229288377,
|
| 1669 |
+
"grad_norm": 26.625,
|
| 1670 |
+
"learning_rate": 1.9602222222222225e-05,
|
| 1671 |
+
"loss": 13.6503,
|
| 1672 |
+
"mean_token_accuracy": 0.7818126812577247,
|
| 1673 |
+
"num_tokens": 7838516.0,
|
| 1674 |
+
"step": 1180
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"entropy": 0.8235707072541117,
|
| 1678 |
+
"epoch": 0.5296834140099037,
|
| 1679 |
+
"grad_norm": 42.0,
|
| 1680 |
+
"learning_rate": 1.9580000000000002e-05,
|
| 1681 |
+
"loss": 13.328,
|
| 1682 |
+
"mean_token_accuracy": 0.7810577813535928,
|
| 1683 |
+
"num_tokens": 7902568.0,
|
| 1684 |
+
"step": 1190
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"entropy": 0.867654961720109,
|
| 1688 |
+
"epoch": 0.5341345351360374,
|
| 1689 |
+
"grad_norm": 29.5,
|
| 1690 |
+
"learning_rate": 1.955777777777778e-05,
|
| 1691 |
+
"loss": 13.7748,
|
| 1692 |
+
"mean_token_accuracy": 0.7776078153401613,
|
| 1693 |
+
"num_tokens": 7969704.0,
|
| 1694 |
+
"step": 1200
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 0.5341345351360374,
|
| 1698 |
+
"eval_biology_entropy": 1.37839106965065,
|
| 1699 |
+
"eval_biology_loss": 1.5514307022094727,
|
| 1700 |
+
"eval_biology_mean_token_accuracy": 0.6508332921862602,
|
| 1701 |
+
"eval_biology_num_tokens": 7969704.0,
|
| 1702 |
+
"eval_biology_runtime": 48.5631,
|
| 1703 |
+
"eval_biology_samples_per_second": 10.296,
|
| 1704 |
+
"eval_biology_steps_per_second": 10.296,
|
| 1705 |
+
"step": 1200
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.5341345351360374,
|
| 1709 |
+
"eval_chemistry_entropy": 0.8148281363546849,
|
| 1710 |
+
"eval_chemistry_loss": 0.8604164719581604,
|
| 1711 |
+
"eval_chemistry_mean_token_accuracy": 0.7772439575791359,
|
| 1712 |
+
"eval_chemistry_num_tokens": 7969704.0,
|
| 1713 |
+
"eval_chemistry_runtime": 56.0492,
|
| 1714 |
+
"eval_chemistry_samples_per_second": 8.921,
|
| 1715 |
+
"eval_chemistry_steps_per_second": 8.921,
|
| 1716 |
+
"step": 1200
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 0.5341345351360374,
|
| 1720 |
+
"eval_math_entropy": 0.7066318797171116,
|
| 1721 |
+
"eval_math_loss": 0.9362333416938782,
|
| 1722 |
+
"eval_math_mean_token_accuracy": 0.7748976874947547,
|
| 1723 |
+
"eval_math_num_tokens": 7969704.0,
|
| 1724 |
+
"eval_math_runtime": 57.7113,
|
| 1725 |
+
"eval_math_samples_per_second": 8.664,
|
| 1726 |
+
"eval_math_steps_per_second": 8.664,
|
| 1727 |
+
"step": 1200
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 0.5341345351360374,
|
| 1731 |
+
"eval_physics_entropy": 0.7838141260743141,
|
| 1732 |
+
"eval_physics_loss": 0.8287554383277893,
|
| 1733 |
+
"eval_physics_mean_token_accuracy": 0.7891261720657349,
|
| 1734 |
+
"eval_physics_num_tokens": 7969704.0,
|
| 1735 |
+
"eval_physics_runtime": 64.5774,
|
| 1736 |
+
"eval_physics_samples_per_second": 7.743,
|
| 1737 |
+
"eval_physics_steps_per_second": 7.743,
|
| 1738 |
+
"step": 1200
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"entropy": 0.823845097143203,
|
| 1742 |
+
"epoch": 0.538585656262171,
|
| 1743 |
+
"grad_norm": 31.375,
|
| 1744 |
+
"learning_rate": 1.9535555555555557e-05,
|
| 1745 |
+
"loss": 13.3304,
|
| 1746 |
+
"mean_token_accuracy": 0.7795415934175253,
|
| 1747 |
+
"num_tokens": 8034765.0,
|
| 1748 |
+
"step": 1210
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"entropy": 0.8282254174351692,
|
| 1752 |
+
"epoch": 0.5430367773883047,
|
| 1753 |
+
"grad_norm": 26.25,
|
| 1754 |
+
"learning_rate": 1.9513333333333335e-05,
|
| 1755 |
+
"loss": 13.8441,
|
| 1756 |
+
"mean_token_accuracy": 0.7799333028495312,
|
| 1757 |
+
"num_tokens": 8103869.0,
|
| 1758 |
+
"step": 1220
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"entropy": 0.8050381439737976,
|
| 1762 |
+
"epoch": 0.5474878985144384,
|
| 1763 |
+
"grad_norm": 28.25,
|
| 1764 |
+
"learning_rate": 1.9491111111111113e-05,
|
| 1765 |
+
"loss": 12.7509,
|
| 1766 |
+
"mean_token_accuracy": 0.7920488726347685,
|
| 1767 |
+
"num_tokens": 8173386.0,
|
| 1768 |
+
"step": 1230
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"entropy": 0.8431567599065601,
|
| 1772 |
+
"epoch": 0.551939019640572,
|
| 1773 |
+
"grad_norm": 27.875,
|
| 1774 |
+
"learning_rate": 1.946888888888889e-05,
|
| 1775 |
+
"loss": 13.6451,
|
| 1776 |
+
"mean_token_accuracy": 0.7757586792111397,
|
| 1777 |
+
"num_tokens": 8242162.0,
|
| 1778 |
+
"step": 1240
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"entropy": 0.8362408608198166,
|
| 1782 |
+
"epoch": 0.5563901407667056,
|
| 1783 |
+
"grad_norm": 32.0,
|
| 1784 |
+
"learning_rate": 1.9446666666666668e-05,
|
| 1785 |
+
"loss": 13.6867,
|
| 1786 |
+
"mean_token_accuracy": 0.7783713828772306,
|
| 1787 |
+
"num_tokens": 8308919.0,
|
| 1788 |
+
"step": 1250
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"entropy": 0.8537883767858148,
|
| 1792 |
+
"epoch": 0.5608412618928392,
|
| 1793 |
+
"grad_norm": 39.5,
|
| 1794 |
+
"learning_rate": 1.9424444444444446e-05,
|
| 1795 |
+
"loss": 14.0357,
|
| 1796 |
+
"mean_token_accuracy": 0.7729044172912836,
|
| 1797 |
+
"num_tokens": 8365617.0,
|
| 1798 |
+
"step": 1260
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"entropy": 0.8582545002922416,
|
| 1802 |
+
"epoch": 0.5652923830189729,
|
| 1803 |
+
"grad_norm": 27.25,
|
| 1804 |
+
"learning_rate": 1.9402222222222223e-05,
|
| 1805 |
+
"loss": 13.7589,
|
| 1806 |
+
"mean_token_accuracy": 0.778778350353241,
|
| 1807 |
+
"num_tokens": 8430728.0,
|
| 1808 |
+
"step": 1270
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"entropy": 0.8236037418246269,
|
| 1812 |
+
"epoch": 0.5697435041451065,
|
| 1813 |
+
"grad_norm": 25.125,
|
| 1814 |
+
"learning_rate": 1.938e-05,
|
| 1815 |
+
"loss": 13.2388,
|
| 1816 |
+
"mean_token_accuracy": 0.784411870315671,
|
| 1817 |
+
"num_tokens": 8497852.0,
|
| 1818 |
+
"step": 1280
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"entropy": 0.8496616246178746,
|
| 1822 |
+
"epoch": 0.5741946252712402,
|
| 1823 |
+
"grad_norm": 27.625,
|
| 1824 |
+
"learning_rate": 1.935777777777778e-05,
|
| 1825 |
+
"loss": 13.9613,
|
| 1826 |
+
"mean_token_accuracy": 0.7750670064240694,
|
| 1827 |
+
"num_tokens": 8558952.0,
|
| 1828 |
+
"step": 1290
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"entropy": 0.8457373520359397,
|
| 1832 |
+
"epoch": 0.5786457463973739,
|
| 1833 |
+
"grad_norm": 32.75,
|
| 1834 |
+
"learning_rate": 1.9335555555555556e-05,
|
| 1835 |
+
"loss": 13.5433,
|
| 1836 |
+
"mean_token_accuracy": 0.7806821152567863,
|
| 1837 |
+
"num_tokens": 8622868.0,
|
| 1838 |
+
"step": 1300
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.5786457463973739,
|
| 1842 |
+
"eval_biology_entropy": 1.43521921479702,
|
| 1843 |
+
"eval_biology_loss": 1.538223147392273,
|
| 1844 |
+
"eval_biology_mean_token_accuracy": 0.6518344734311103,
|
| 1845 |
+
"eval_biology_num_tokens": 8622868.0,
|
| 1846 |
+
"eval_biology_runtime": 48.4265,
|
| 1847 |
+
"eval_biology_samples_per_second": 10.325,
|
| 1848 |
+
"eval_biology_steps_per_second": 10.325,
|
| 1849 |
+
"step": 1300
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.5786457463973739,
|
| 1853 |
+
"eval_chemistry_entropy": 0.8178334766924381,
|
| 1854 |
+
"eval_chemistry_loss": 0.849298357963562,
|
| 1855 |
+
"eval_chemistry_mean_token_accuracy": 0.7795388196110725,
|
| 1856 |
+
"eval_chemistry_num_tokens": 8622868.0,
|
| 1857 |
+
"eval_chemistry_runtime": 56.0158,
|
| 1858 |
+
"eval_chemistry_samples_per_second": 8.926,
|
| 1859 |
+
"eval_chemistry_steps_per_second": 8.926,
|
| 1860 |
+
"step": 1300
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 0.5786457463973739,
|
| 1864 |
+
"eval_math_entropy": 0.705706961274147,
|
| 1865 |
+
"eval_math_loss": 0.9391148686408997,
|
| 1866 |
+
"eval_math_mean_token_accuracy": 0.7748868613243103,
|
| 1867 |
+
"eval_math_num_tokens": 8622868.0,
|
| 1868 |
+
"eval_math_runtime": 57.6077,
|
| 1869 |
+
"eval_math_samples_per_second": 8.679,
|
| 1870 |
+
"eval_math_steps_per_second": 8.679,
|
| 1871 |
+
"step": 1300
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.5786457463973739,
|
| 1875 |
+
"eval_physics_entropy": 0.770459543287754,
|
| 1876 |
+
"eval_physics_loss": 0.8210808038711548,
|
| 1877 |
+
"eval_physics_mean_token_accuracy": 0.7902451878190041,
|
| 1878 |
+
"eval_physics_num_tokens": 8622868.0,
|
| 1879 |
+
"eval_physics_runtime": 64.5139,
|
| 1880 |
+
"eval_physics_samples_per_second": 7.75,
|
| 1881 |
+
"eval_physics_steps_per_second": 7.75,
|
| 1882 |
+
"step": 1300
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"entropy": 0.8071680258028209,
|
| 1886 |
+
"epoch": 0.5830968675235075,
|
| 1887 |
+
"grad_norm": 32.5,
|
| 1888 |
+
"learning_rate": 1.9313333333333334e-05,
|
| 1889 |
+
"loss": 13.2153,
|
| 1890 |
+
"mean_token_accuracy": 0.7843220077455044,
|
| 1891 |
+
"num_tokens": 8694510.0,
|
| 1892 |
+
"step": 1310
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"entropy": 0.8942232267931104,
|
| 1896 |
+
"epoch": 0.5875479886496411,
|
| 1897 |
+
"grad_norm": 33.5,
|
| 1898 |
+
"learning_rate": 1.9291111111111115e-05,
|
| 1899 |
+
"loss": 14.6327,
|
| 1900 |
+
"mean_token_accuracy": 0.7653406778350472,
|
| 1901 |
+
"num_tokens": 8757753.0,
|
| 1902 |
+
"step": 1320
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"entropy": 0.8597627034410834,
|
| 1906 |
+
"epoch": 0.5919991097757747,
|
| 1907 |
+
"grad_norm": 30.75,
|
| 1908 |
+
"learning_rate": 1.926888888888889e-05,
|
| 1909 |
+
"loss": 13.8741,
|
| 1910 |
+
"mean_token_accuracy": 0.7796951614320278,
|
| 1911 |
+
"num_tokens": 8823153.0,
|
| 1912 |
+
"step": 1330
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"entropy": 0.8664615718647838,
|
| 1916 |
+
"epoch": 0.5964502309019084,
|
| 1917 |
+
"grad_norm": 27.75,
|
| 1918 |
+
"learning_rate": 1.924666666666667e-05,
|
| 1919 |
+
"loss": 14.0396,
|
| 1920 |
+
"mean_token_accuracy": 0.7731820710003376,
|
| 1921 |
+
"num_tokens": 8886401.0,
|
| 1922 |
+
"step": 1340
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"entropy": 0.8158455387689173,
|
| 1926 |
+
"epoch": 0.6009013520280421,
|
| 1927 |
+
"grad_norm": 25.0,
|
| 1928 |
+
"learning_rate": 1.9224444444444444e-05,
|
| 1929 |
+
"loss": 13.1708,
|
| 1930 |
+
"mean_token_accuracy": 0.7833993557840586,
|
| 1931 |
+
"num_tokens": 8958530.0,
|
| 1932 |
+
"step": 1350
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"entropy": 0.8429827799089253,
|
| 1936 |
+
"epoch": 0.6053524731541757,
|
| 1937 |
+
"grad_norm": 29.5,
|
| 1938 |
+
"learning_rate": 1.9202222222222225e-05,
|
| 1939 |
+
"loss": 13.6607,
|
| 1940 |
+
"mean_token_accuracy": 0.7796129040420056,
|
| 1941 |
+
"num_tokens": 9024677.0,
|
| 1942 |
+
"step": 1360
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"entropy": 0.8648322455585002,
|
| 1946 |
+
"epoch": 0.6098035942803094,
|
| 1947 |
+
"grad_norm": 30.75,
|
| 1948 |
+
"learning_rate": 1.918e-05,
|
| 1949 |
+
"loss": 14.0503,
|
| 1950 |
+
"mean_token_accuracy": 0.7712652169167995,
|
| 1951 |
+
"num_tokens": 9090280.0,
|
| 1952 |
+
"step": 1370
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"entropy": 0.8274468386545777,
|
| 1956 |
+
"epoch": 0.614254715406443,
|
| 1957 |
+
"grad_norm": 31.0,
|
| 1958 |
+
"learning_rate": 1.915777777777778e-05,
|
| 1959 |
+
"loss": 13.5605,
|
| 1960 |
+
"mean_token_accuracy": 0.7780856698751449,
|
| 1961 |
+
"num_tokens": 9153771.0,
|
| 1962 |
+
"step": 1380
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"entropy": 0.7939455186016857,
|
| 1966 |
+
"epoch": 0.6187058365325766,
|
| 1967 |
+
"grad_norm": 28.5,
|
| 1968 |
+
"learning_rate": 1.9135555555555555e-05,
|
| 1969 |
+
"loss": 12.6986,
|
| 1970 |
+
"mean_token_accuracy": 0.7918769404292106,
|
| 1971 |
+
"num_tokens": 9227117.0,
|
| 1972 |
+
"step": 1390
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"entropy": 0.8405648414045572,
|
| 1976 |
+
"epoch": 0.6231569576587103,
|
| 1977 |
+
"grad_norm": 31.25,
|
| 1978 |
+
"learning_rate": 1.9113333333333336e-05,
|
| 1979 |
+
"loss": 13.9305,
|
| 1980 |
+
"mean_token_accuracy": 0.7758563198149204,
|
| 1981 |
+
"num_tokens": 9291760.0,
|
| 1982 |
+
"step": 1400
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.6231569576587103,
|
| 1986 |
+
"eval_biology_entropy": 1.4560990427136422,
|
| 1987 |
+
"eval_biology_loss": 1.5385361909866333,
|
| 1988 |
+
"eval_biology_mean_token_accuracy": 0.6521896251440048,
|
| 1989 |
+
"eval_biology_num_tokens": 9291760.0,
|
| 1990 |
+
"eval_biology_runtime": 48.3888,
|
| 1991 |
+
"eval_biology_samples_per_second": 10.333,
|
| 1992 |
+
"eval_biology_steps_per_second": 10.333,
|
| 1993 |
+
"step": 1400
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 0.6231569576587103,
|
| 1997 |
+
"eval_chemistry_entropy": 0.8454484658241272,
|
| 1998 |
+
"eval_chemistry_loss": 0.8420565724372864,
|
| 1999 |
+
"eval_chemistry_mean_token_accuracy": 0.7808461990356446,
|
| 2000 |
+
"eval_chemistry_num_tokens": 9291760.0,
|
| 2001 |
+
"eval_chemistry_runtime": 55.9558,
|
| 2002 |
+
"eval_chemistry_samples_per_second": 8.936,
|
| 2003 |
+
"eval_chemistry_steps_per_second": 8.936,
|
| 2004 |
+
"step": 1400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.6231569576587103,
|
| 2008 |
+
"eval_math_entropy": 0.7586602506041527,
|
| 2009 |
+
"eval_math_loss": 0.9284656643867493,
|
| 2010 |
+
"eval_math_mean_token_accuracy": 0.7747275605797768,
|
| 2011 |
+
"eval_math_num_tokens": 9291760.0,
|
| 2012 |
+
"eval_math_runtime": 57.6406,
|
| 2013 |
+
"eval_math_samples_per_second": 8.674,
|
| 2014 |
+
"eval_math_steps_per_second": 8.674,
|
| 2015 |
+
"step": 1400
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 0.6231569576587103,
|
| 2019 |
+
"eval_physics_entropy": 0.8177245287299156,
|
| 2020 |
+
"eval_physics_loss": 0.8168751001358032,
|
| 2021 |
+
"eval_physics_mean_token_accuracy": 0.7908310489654541,
|
| 2022 |
+
"eval_physics_num_tokens": 9291760.0,
|
| 2023 |
+
"eval_physics_runtime": 64.5434,
|
| 2024 |
+
"eval_physics_samples_per_second": 7.747,
|
| 2025 |
+
"eval_physics_steps_per_second": 7.747,
|
| 2026 |
+
"step": 1400
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"entropy": 0.8965679749846458,
|
| 2030 |
+
"epoch": 0.6276080787848439,
|
| 2031 |
+
"grad_norm": 28.875,
|
| 2032 |
+
"learning_rate": 1.9091111111111113e-05,
|
| 2033 |
+
"loss": 14.1354,
|
| 2034 |
+
"mean_token_accuracy": 0.7691367525607348,
|
| 2035 |
+
"num_tokens": 9357561.0,
|
| 2036 |
+
"step": 1410
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"entropy": 0.8075421938672662,
|
| 2040 |
+
"epoch": 0.6320591999109776,
|
| 2041 |
+
"grad_norm": 36.25,
|
| 2042 |
+
"learning_rate": 1.906888888888889e-05,
|
| 2043 |
+
"loss": 13.3559,
|
| 2044 |
+
"mean_token_accuracy": 0.7846433013677597,
|
| 2045 |
+
"num_tokens": 9425022.0,
|
| 2046 |
+
"step": 1420
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"entropy": 0.8154678157530725,
|
| 2050 |
+
"epoch": 0.6365103210371112,
|
| 2051 |
+
"grad_norm": 30.75,
|
| 2052 |
+
"learning_rate": 1.904666666666667e-05,
|
| 2053 |
+
"loss": 13.1969,
|
| 2054 |
+
"mean_token_accuracy": 0.7874299377202988,
|
| 2055 |
+
"num_tokens": 9492028.0,
|
| 2056 |
+
"step": 1430
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"entropy": 0.795359942317009,
|
| 2060 |
+
"epoch": 0.6409614421632449,
|
| 2061 |
+
"grad_norm": 33.5,
|
| 2062 |
+
"learning_rate": 1.9024444444444446e-05,
|
| 2063 |
+
"loss": 12.7361,
|
| 2064 |
+
"mean_token_accuracy": 0.7892100885510445,
|
| 2065 |
+
"num_tokens": 9561091.0,
|
| 2066 |
+
"step": 1440
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"entropy": 0.8082486374303699,
|
| 2070 |
+
"epoch": 0.6454125632893785,
|
| 2071 |
+
"grad_norm": 36.25,
|
| 2072 |
+
"learning_rate": 1.9002222222222224e-05,
|
| 2073 |
+
"loss": 13.4226,
|
| 2074 |
+
"mean_token_accuracy": 0.7825187552720309,
|
| 2075 |
+
"num_tokens": 9628383.0,
|
| 2076 |
+
"step": 1450
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"entropy": 0.7709562292322516,
|
| 2080 |
+
"epoch": 0.6498636844155121,
|
| 2081 |
+
"grad_norm": 24.875,
|
| 2082 |
+
"learning_rate": 1.898e-05,
|
| 2083 |
+
"loss": 12.3092,
|
| 2084 |
+
"mean_token_accuracy": 0.7987042464315891,
|
| 2085 |
+
"num_tokens": 9695723.0,
|
| 2086 |
+
"step": 1460
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"entropy": 0.8008807510137558,
|
| 2090 |
+
"epoch": 0.6543148055416458,
|
| 2091 |
+
"grad_norm": 27.75,
|
| 2092 |
+
"learning_rate": 1.895777777777778e-05,
|
| 2093 |
+
"loss": 13.1207,
|
| 2094 |
+
"mean_token_accuracy": 0.7885788295418024,
|
| 2095 |
+
"num_tokens": 9763393.0,
|
| 2096 |
+
"step": 1470
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"entropy": 0.8602127678692341,
|
| 2100 |
+
"epoch": 0.6587659266677794,
|
| 2101 |
+
"grad_norm": 30.625,
|
| 2102 |
+
"learning_rate": 1.8935555555555556e-05,
|
| 2103 |
+
"loss": 13.8264,
|
| 2104 |
+
"mean_token_accuracy": 0.7781572956591845,
|
| 2105 |
+
"num_tokens": 9827272.0,
|
| 2106 |
+
"step": 1480
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"entropy": 0.8142364447936415,
|
| 2110 |
+
"epoch": 0.6632170477939131,
|
| 2111 |
+
"grad_norm": 25.75,
|
| 2112 |
+
"learning_rate": 1.8913333333333334e-05,
|
| 2113 |
+
"loss": 13.1217,
|
| 2114 |
+
"mean_token_accuracy": 0.783425547555089,
|
| 2115 |
+
"num_tokens": 9892418.0,
|
| 2116 |
+
"step": 1490
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"entropy": 0.7871226992458105,
|
| 2120 |
+
"epoch": 0.6676681689200468,
|
| 2121 |
+
"grad_norm": 30.25,
|
| 2122 |
+
"learning_rate": 1.8891111111111115e-05,
|
| 2123 |
+
"loss": 12.9626,
|
| 2124 |
+
"mean_token_accuracy": 0.7883934937417507,
|
| 2125 |
+
"num_tokens": 9955369.0,
|
| 2126 |
+
"step": 1500
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 0.6676681689200468,
|
| 2130 |
+
"eval_biology_entropy": 1.4557775322794915,
|
| 2131 |
+
"eval_biology_loss": 1.5257188081741333,
|
| 2132 |
+
"eval_biology_mean_token_accuracy": 0.6541550720334053,
|
| 2133 |
+
"eval_biology_num_tokens": 9955369.0,
|
| 2134 |
+
"eval_biology_runtime": 48.3598,
|
| 2135 |
+
"eval_biology_samples_per_second": 10.339,
|
| 2136 |
+
"eval_biology_steps_per_second": 10.339,
|
| 2137 |
+
"step": 1500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.6676681689200468,
|
| 2141 |
+
"eval_chemistry_entropy": 0.8462375261187554,
|
| 2142 |
+
"eval_chemistry_loss": 0.8326632976531982,
|
| 2143 |
+
"eval_chemistry_mean_token_accuracy": 0.7831478416323662,
|
| 2144 |
+
"eval_chemistry_num_tokens": 9955369.0,
|
| 2145 |
+
"eval_chemistry_runtime": 56.166,
|
| 2146 |
+
"eval_chemistry_samples_per_second": 8.902,
|
| 2147 |
+
"eval_chemistry_steps_per_second": 8.902,
|
| 2148 |
+
"step": 1500
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 0.6676681689200468,
|
| 2152 |
+
"eval_math_entropy": 0.7239785996079445,
|
| 2153 |
+
"eval_math_loss": 0.9294689893722534,
|
| 2154 |
+
"eval_math_mean_token_accuracy": 0.7758806052207947,
|
| 2155 |
+
"eval_math_num_tokens": 9955369.0,
|
| 2156 |
+
"eval_math_runtime": 57.7436,
|
| 2157 |
+
"eval_math_samples_per_second": 8.659,
|
| 2158 |
+
"eval_math_steps_per_second": 8.659,
|
| 2159 |
+
"step": 1500
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.6676681689200468,
|
| 2163 |
+
"eval_physics_entropy": 0.80675083309412,
|
| 2164 |
+
"eval_physics_loss": 0.8114446997642517,
|
| 2165 |
+
"eval_physics_mean_token_accuracy": 0.7917336781024933,
|
| 2166 |
+
"eval_physics_num_tokens": 9955369.0,
|
| 2167 |
+
"eval_physics_runtime": 64.6166,
|
| 2168 |
+
"eval_physics_samples_per_second": 7.738,
|
| 2169 |
+
"eval_physics_steps_per_second": 7.738,
|
| 2170 |
+
"step": 1500
|
| 2171 |
+
}
|
| 2172 |
+
],
|
| 2173 |
+
"logging_steps": 10,
|
| 2174 |
+
"max_steps": 10000,
|
| 2175 |
+
"num_input_tokens_seen": 0,
|
| 2176 |
+
"num_train_epochs": 5,
|
| 2177 |
+
"save_steps": 500,
|
| 2178 |
+
"stateful_callbacks": {
|
| 2179 |
+
"TrainerControl": {
|
| 2180 |
+
"args": {
|
| 2181 |
+
"should_epoch_stop": false,
|
| 2182 |
+
"should_evaluate": false,
|
| 2183 |
+
"should_log": false,
|
| 2184 |
+
"should_save": true,
|
| 2185 |
+
"should_training_stop": false
|
| 2186 |
+
},
|
| 2187 |
+
"attributes": {}
|
| 2188 |
+
}
|
| 2189 |
+
},
|
| 2190 |
+
"total_flos": 6.6755096025463e+17,
|
| 2191 |
+
"train_batch_size": 1,
|
| 2192 |
+
"trial_name": null,
|
| 2193 |
+
"trial_params": null
|
| 2194 |
+
}
|
checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b3edfed8df0c4397da38d1f8b970fbbf2479f63c7874fb7f01978dcc690947a
|
| 3 |
+
size 6481
|