Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- checkpoint-2000/added_tokens.json +3 -0
- checkpoint-2000/chat_template.jinja +47 -0
- checkpoint-2000/config.json +110 -0
- checkpoint-2000/generation_config.json +13 -0
- checkpoint-2000/model-00001-of-00005.safetensors +3 -0
- checkpoint-2000/model-00002-of-00005.safetensors +3 -0
- checkpoint-2000/model-00003-of-00005.safetensors +3 -0
- checkpoint-2000/model-00004-of-00005.safetensors +3 -0
- checkpoint-2000/model-00005-of-00005.safetensors +3 -0
- checkpoint-2000/model.safetensors.index.json +0 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/special_tokens_map.json +33 -0
- checkpoint-2000/tokenizer.json +3 -0
- checkpoint-2000/tokenizer.model +3 -0
- checkpoint-2000/tokenizer_config.json +0 -0
- checkpoint-2000/trainer_state.json +2914 -0
- checkpoint-2000/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-2000/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image_soft_token>": 262144
|
| 3 |
+
}
|
checkpoint-2000/chat_template.jinja
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{{ bos_token }}
|
| 2 |
+
{%- if messages[0]['role'] == 'system' -%}
|
| 3 |
+
{%- if messages[0]['content'] is string -%}
|
| 4 |
+
{%- set first_user_prefix = messages[0]['content'] + '
|
| 5 |
+
|
| 6 |
+
' -%}
|
| 7 |
+
{%- else -%}
|
| 8 |
+
{%- set first_user_prefix = messages[0]['content'][0]['text'] + '
|
| 9 |
+
|
| 10 |
+
' -%}
|
| 11 |
+
{%- endif -%}
|
| 12 |
+
{%- set loop_messages = messages[1:] -%}
|
| 13 |
+
{%- else -%}
|
| 14 |
+
{%- set first_user_prefix = "" -%}
|
| 15 |
+
{%- set loop_messages = messages -%}
|
| 16 |
+
{%- endif -%}
|
| 17 |
+
{%- for message in loop_messages -%}
|
| 18 |
+
{%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
|
| 19 |
+
{{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
|
| 20 |
+
{%- endif -%}
|
| 21 |
+
{%- if (message['role'] == 'assistant') -%}
|
| 22 |
+
{%- set role = "model" -%}
|
| 23 |
+
{%- else -%}
|
| 24 |
+
{%- set role = message['role'] -%}
|
| 25 |
+
{%- endif -%}
|
| 26 |
+
{{ '<start_of_turn>' + role + '
|
| 27 |
+
' + (first_user_prefix if loop.first else "") }}
|
| 28 |
+
{%- if message['content'] is string -%}
|
| 29 |
+
{{ message['content'] | trim }}
|
| 30 |
+
{%- elif message['content'] is iterable -%}
|
| 31 |
+
{%- for item in message['content'] -%}
|
| 32 |
+
{%- if item['type'] == 'image' -%}
|
| 33 |
+
{{ '<start_of_image>' }}
|
| 34 |
+
{%- elif item['type'] == 'text' -%}
|
| 35 |
+
{{ item['text'] | trim }}
|
| 36 |
+
{%- endif -%}
|
| 37 |
+
{%- endfor -%}
|
| 38 |
+
{%- else -%}
|
| 39 |
+
{{ raise_exception("Invalid content type") }}
|
| 40 |
+
{%- endif -%}
|
| 41 |
+
{{ '<end_of_turn>
|
| 42 |
+
' }}
|
| 43 |
+
{%- endfor -%}
|
| 44 |
+
{%- if add_generation_prompt -%}
|
| 45 |
+
{{'<start_of_turn>model
|
| 46 |
+
'}}
|
| 47 |
+
{%- endif -%}
|
checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma3ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"boi_token_index": 255999,
|
| 6 |
+
"bos_token_id": 2,
|
| 7 |
+
"dtype": "bfloat16",
|
| 8 |
+
"eoi_token_index": 256000,
|
| 9 |
+
"eos_token_id": 1,
|
| 10 |
+
"image_token_index": 262144,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"mm_tokens_per_image": 256,
|
| 13 |
+
"model_type": "gemma3",
|
| 14 |
+
"pad_token_id": 0,
|
| 15 |
+
"text_config": {
|
| 16 |
+
"_sliding_window_pattern": 6,
|
| 17 |
+
"attention_bias": false,
|
| 18 |
+
"attention_dropout": 0.0,
|
| 19 |
+
"attn_logit_softcapping": null,
|
| 20 |
+
"dtype": "bfloat16",
|
| 21 |
+
"final_logit_softcapping": null,
|
| 22 |
+
"head_dim": 256,
|
| 23 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 24 |
+
"hidden_size": 3840,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"intermediate_size": 15360,
|
| 27 |
+
"layer_types": [
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"sliding_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"sliding_attention",
|
| 38 |
+
"sliding_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"sliding_attention",
|
| 41 |
+
"sliding_attention",
|
| 42 |
+
"sliding_attention",
|
| 43 |
+
"sliding_attention",
|
| 44 |
+
"sliding_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"sliding_attention",
|
| 47 |
+
"sliding_attention",
|
| 48 |
+
"sliding_attention",
|
| 49 |
+
"sliding_attention",
|
| 50 |
+
"sliding_attention",
|
| 51 |
+
"full_attention",
|
| 52 |
+
"sliding_attention",
|
| 53 |
+
"sliding_attention",
|
| 54 |
+
"sliding_attention",
|
| 55 |
+
"sliding_attention",
|
| 56 |
+
"sliding_attention",
|
| 57 |
+
"full_attention",
|
| 58 |
+
"sliding_attention",
|
| 59 |
+
"sliding_attention",
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"sliding_attention",
|
| 63 |
+
"full_attention",
|
| 64 |
+
"sliding_attention",
|
| 65 |
+
"sliding_attention",
|
| 66 |
+
"sliding_attention",
|
| 67 |
+
"sliding_attention",
|
| 68 |
+
"sliding_attention",
|
| 69 |
+
"full_attention",
|
| 70 |
+
"sliding_attention",
|
| 71 |
+
"sliding_attention",
|
| 72 |
+
"sliding_attention",
|
| 73 |
+
"sliding_attention",
|
| 74 |
+
"sliding_attention",
|
| 75 |
+
"full_attention"
|
| 76 |
+
],
|
| 77 |
+
"max_position_embeddings": 131072,
|
| 78 |
+
"model_type": "gemma3_text",
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_hidden_layers": 48,
|
| 81 |
+
"num_key_value_heads": 8,
|
| 82 |
+
"query_pre_attn_scalar": 256,
|
| 83 |
+
"rms_norm_eps": 1e-06,
|
| 84 |
+
"rope_local_base_freq": 10000.0,
|
| 85 |
+
"rope_scaling": {
|
| 86 |
+
"factor": 8.0,
|
| 87 |
+
"rope_type": "linear"
|
| 88 |
+
},
|
| 89 |
+
"rope_theta": 1000000.0,
|
| 90 |
+
"sliding_window": 1024,
|
| 91 |
+
"use_cache": true,
|
| 92 |
+
"vocab_size": 262208
|
| 93 |
+
},
|
| 94 |
+
"transformers_version": "4.56.1",
|
| 95 |
+
"vision_config": {
|
| 96 |
+
"attention_dropout": 0.0,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"image_size": 896,
|
| 101 |
+
"intermediate_size": 4304,
|
| 102 |
+
"layer_norm_eps": 1e-06,
|
| 103 |
+
"model_type": "siglip_vision_model",
|
| 104 |
+
"num_attention_heads": 16,
|
| 105 |
+
"num_channels": 3,
|
| 106 |
+
"num_hidden_layers": 27,
|
| 107 |
+
"patch_size": 14,
|
| 108 |
+
"vision_use_head": false
|
| 109 |
+
}
|
| 110 |
+
}
|
checkpoint-2000/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 2,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
1,
|
| 6 |
+
1,
|
| 7 |
+
106
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 0,
|
| 10 |
+
"top_k": 64,
|
| 11 |
+
"top_p": 0.95,
|
| 12 |
+
"transformers_version": "4.56.1"
|
| 13 |
+
}
|
checkpoint-2000/model-00001-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4847447e92599833e8dbaa3067cd201c3bb5c052efa91f11ba891e43234f7832
|
| 3 |
+
size 4979902192
|
checkpoint-2000/model-00002-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:891bd54eed03cba9ee1e705533a02a8217fcc29f356e4a1f53e5fd0d178883ad
|
| 3 |
+
size 4931296592
|
checkpoint-2000/model-00003-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
|
| 3 |
+
size 4931296656
|
checkpoint-2000/model-00004-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
|
| 3 |
+
size 4931296656
|
checkpoint-2000/model-00005-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:16da08a7f30ca187fb1a5d3c8dfc76d76e487910aa1b113bc513a52ab2a41114
|
| 3 |
+
size 4601000928
|
checkpoint-2000/model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:931873b56ccf0c294ce5083f41400b9a20ac4ce03abc12b340488007a9347a15
|
| 3 |
+
size 896607944
|
checkpoint-2000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4132949318e028debc59f4db164bc9a537cbb130ee35d892cb08945d6ab6790a
|
| 3 |
+
size 14645
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c56cc7d2ab3dfe9edf7a28cb038a85004572c59f841d9bdfd8a137d5d874b19
|
| 3 |
+
size 1465
|
checkpoint-2000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"boi_token": "<start_of_image>",
|
| 3 |
+
"bos_token": {
|
| 4 |
+
"content": "<bos>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
"eoi_token": "<end_of_image>",
|
| 11 |
+
"eos_token": {
|
| 12 |
+
"content": "<eos>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"image_token": "<image_soft_token>",
|
| 19 |
+
"pad_token": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false
|
| 25 |
+
},
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"content": "<unk>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": false,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
}
|
| 33 |
+
}
|
checkpoint-2000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
|
| 3 |
+
size 33384568
|
checkpoint-2000/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
|
| 3 |
+
size 4689074
|
checkpoint-2000/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,2914 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.2,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 2000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 0.7081545192748309,
|
| 14 |
+
"epoch": 0.016,
|
| 15 |
+
"grad_norm": 119.5,
|
| 16 |
+
"learning_rate": 6.000000000000001e-07,
|
| 17 |
+
"loss": 30.4211,
|
| 18 |
+
"mean_token_accuracy": 0.7030085142701864,
|
| 19 |
+
"num_tokens": 338352.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 0.7043411549180746,
|
| 24 |
+
"epoch": 0.032,
|
| 25 |
+
"grad_norm": 111.5,
|
| 26 |
+
"learning_rate": 1.2666666666666669e-06,
|
| 27 |
+
"loss": 30.1214,
|
| 28 |
+
"mean_token_accuracy": 0.7080705553293228,
|
| 29 |
+
"num_tokens": 671193.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 0.7044402219355106,
|
| 34 |
+
"epoch": 0.048,
|
| 35 |
+
"grad_norm": 106.0,
|
| 36 |
+
"learning_rate": 1.9333333333333336e-06,
|
| 37 |
+
"loss": 29.7341,
|
| 38 |
+
"mean_token_accuracy": 0.7064134150743484,
|
| 39 |
+
"num_tokens": 1001008.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 0.733870861120522,
|
| 44 |
+
"epoch": 0.064,
|
| 45 |
+
"grad_norm": 96.0,
|
| 46 |
+
"learning_rate": 2.6e-06,
|
| 47 |
+
"loss": 30.1494,
|
| 48 |
+
"mean_token_accuracy": 0.7005880128592252,
|
| 49 |
+
"num_tokens": 1331208.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 0.7398215295746923,
|
| 54 |
+
"epoch": 0.08,
|
| 55 |
+
"grad_norm": 90.5,
|
| 56 |
+
"learning_rate": 3.266666666666667e-06,
|
| 57 |
+
"loss": 28.7202,
|
| 58 |
+
"mean_token_accuracy": 0.7051243670284748,
|
| 59 |
+
"num_tokens": 1667532.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 0.738157893717289,
|
| 64 |
+
"epoch": 0.096,
|
| 65 |
+
"grad_norm": 76.5,
|
| 66 |
+
"learning_rate": 3.9333333333333335e-06,
|
| 67 |
+
"loss": 27.1633,
|
| 68 |
+
"mean_token_accuracy": 0.71303277797997,
|
| 69 |
+
"num_tokens": 2007176.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 0.8131938450038433,
|
| 74 |
+
"epoch": 0.112,
|
| 75 |
+
"grad_norm": 57.5,
|
| 76 |
+
"learning_rate": 4.600000000000001e-06,
|
| 77 |
+
"loss": 26.4428,
|
| 78 |
+
"mean_token_accuracy": 0.7120848346501589,
|
| 79 |
+
"num_tokens": 2342259.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 0.9618701986968518,
|
| 84 |
+
"epoch": 0.128,
|
| 85 |
+
"grad_norm": 43.75,
|
| 86 |
+
"learning_rate": 5.2666666666666665e-06,
|
| 87 |
+
"loss": 25.9908,
|
| 88 |
+
"mean_token_accuracy": 0.7077770136296749,
|
| 89 |
+
"num_tokens": 2663987.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.0031472396105527,
|
| 94 |
+
"epoch": 0.144,
|
| 95 |
+
"grad_norm": 30.125,
|
| 96 |
+
"learning_rate": 5.933333333333335e-06,
|
| 97 |
+
"loss": 23.3524,
|
| 98 |
+
"mean_token_accuracy": 0.7268724206835031,
|
| 99 |
+
"num_tokens": 2997824.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.0786659345030785,
|
| 104 |
+
"epoch": 0.16,
|
| 105 |
+
"grad_norm": 20.75,
|
| 106 |
+
"learning_rate": 6.600000000000001e-06,
|
| 107 |
+
"loss": 21.8746,
|
| 108 |
+
"mean_token_accuracy": 0.7358953565359115,
|
| 109 |
+
"num_tokens": 3330597.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.16,
|
| 114 |
+
"eval_biology_entropy": 3.260982286453247,
|
| 115 |
+
"eval_biology_loss": 3.7093346118927,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.46263515305519104,
|
| 117 |
+
"eval_biology_num_tokens": 3330597.0,
|
| 118 |
+
"eval_biology_runtime": 57.0058,
|
| 119 |
+
"eval_biology_samples_per_second": 8.771,
|
| 120 |
+
"eval_biology_steps_per_second": 2.193,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.16,
|
| 125 |
+
"eval_chemistry_entropy": 1.6407185397148132,
|
| 126 |
+
"eval_chemistry_loss": 1.7927597761154175,
|
| 127 |
+
"eval_chemistry_mean_token_accuracy": 0.677183347940445,
|
| 128 |
+
"eval_chemistry_num_tokens": 3330597.0,
|
| 129 |
+
"eval_chemistry_runtime": 71.6705,
|
| 130 |
+
"eval_chemistry_samples_per_second": 6.976,
|
| 131 |
+
"eval_chemistry_steps_per_second": 1.744,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.16,
|
| 136 |
+
"eval_math_entropy": 0.9620108480453491,
|
| 137 |
+
"eval_math_loss": 1.4342334270477295,
|
| 138 |
+
"eval_math_mean_token_accuracy": 0.7288489332199096,
|
| 139 |
+
"eval_math_num_tokens": 3330597.0,
|
| 140 |
+
"eval_math_runtime": 74.2141,
|
| 141 |
+
"eval_math_samples_per_second": 6.737,
|
| 142 |
+
"eval_math_steps_per_second": 1.684,
|
| 143 |
+
"step": 100
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16,
|
| 147 |
+
"eval_physics_entropy": 1.1280243759155273,
|
| 148 |
+
"eval_physics_loss": 1.3446351289749146,
|
| 149 |
+
"eval_physics_mean_token_accuracy": 0.7332401766777038,
|
| 150 |
+
"eval_physics_num_tokens": 3330597.0,
|
| 151 |
+
"eval_physics_runtime": 84.9827,
|
| 152 |
+
"eval_physics_samples_per_second": 5.884,
|
| 153 |
+
"eval_physics_steps_per_second": 1.471,
|
| 154 |
+
"step": 100
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"entropy": 1.1533773414790631,
|
| 158 |
+
"epoch": 0.176,
|
| 159 |
+
"grad_norm": 19.75,
|
| 160 |
+
"learning_rate": 7.266666666666668e-06,
|
| 161 |
+
"loss": 21.3285,
|
| 162 |
+
"mean_token_accuracy": 0.7356826655566693,
|
| 163 |
+
"num_tokens": 3658264.0,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"entropy": 1.112282781302929,
|
| 168 |
+
"epoch": 0.192,
|
| 169 |
+
"grad_norm": 12.625,
|
| 170 |
+
"learning_rate": 7.933333333333334e-06,
|
| 171 |
+
"loss": 20.4084,
|
| 172 |
+
"mean_token_accuracy": 0.7426700782030821,
|
| 173 |
+
"num_tokens": 3995568.0,
|
| 174 |
+
"step": 120
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"entropy": 1.086438114568591,
|
| 178 |
+
"epoch": 0.208,
|
| 179 |
+
"grad_norm": 13.875,
|
| 180 |
+
"learning_rate": 8.6e-06,
|
| 181 |
+
"loss": 19.3962,
|
| 182 |
+
"mean_token_accuracy": 0.7525120593607426,
|
| 183 |
+
"num_tokens": 4321436.0,
|
| 184 |
+
"step": 130
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"entropy": 1.089601282030344,
|
| 188 |
+
"epoch": 0.224,
|
| 189 |
+
"grad_norm": 11.8125,
|
| 190 |
+
"learning_rate": 9.266666666666667e-06,
|
| 191 |
+
"loss": 19.0472,
|
| 192 |
+
"mean_token_accuracy": 0.7530231148004531,
|
| 193 |
+
"num_tokens": 4648491.0,
|
| 194 |
+
"step": 140
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"entropy": 1.0804125092923642,
|
| 198 |
+
"epoch": 0.24,
|
| 199 |
+
"grad_norm": 11.8125,
|
| 200 |
+
"learning_rate": 9.933333333333334e-06,
|
| 201 |
+
"loss": 18.7739,
|
| 202 |
+
"mean_token_accuracy": 0.7554640270769596,
|
| 203 |
+
"num_tokens": 4986175.0,
|
| 204 |
+
"step": 150
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"entropy": 1.0836318269371987,
|
| 208 |
+
"epoch": 0.256,
|
| 209 |
+
"grad_norm": 11.9375,
|
| 210 |
+
"learning_rate": 1.0600000000000002e-05,
|
| 211 |
+
"loss": 18.4704,
|
| 212 |
+
"mean_token_accuracy": 0.7584241010248661,
|
| 213 |
+
"num_tokens": 5329320.0,
|
| 214 |
+
"step": 160
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"entropy": 1.0882270745933056,
|
| 218 |
+
"epoch": 0.272,
|
| 219 |
+
"grad_norm": 13.375,
|
| 220 |
+
"learning_rate": 1.1266666666666668e-05,
|
| 221 |
+
"loss": 18.6237,
|
| 222 |
+
"mean_token_accuracy": 0.754904105886817,
|
| 223 |
+
"num_tokens": 5658796.0,
|
| 224 |
+
"step": 170
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"entropy": 1.076054659858346,
|
| 228 |
+
"epoch": 0.288,
|
| 229 |
+
"grad_norm": 11.0625,
|
| 230 |
+
"learning_rate": 1.1933333333333335e-05,
|
| 231 |
+
"loss": 18.4139,
|
| 232 |
+
"mean_token_accuracy": 0.7581049785017967,
|
| 233 |
+
"num_tokens": 5980201.0,
|
| 234 |
+
"step": 180
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"entropy": 1.072243456915021,
|
| 238 |
+
"epoch": 0.304,
|
| 239 |
+
"grad_norm": 15.5,
|
| 240 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 241 |
+
"loss": 18.1532,
|
| 242 |
+
"mean_token_accuracy": 0.7589787419885397,
|
| 243 |
+
"num_tokens": 6298858.0,
|
| 244 |
+
"step": 190
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"entropy": 1.0509168311953545,
|
| 248 |
+
"epoch": 0.32,
|
| 249 |
+
"grad_norm": 12.0,
|
| 250 |
+
"learning_rate": 1.3266666666666668e-05,
|
| 251 |
+
"loss": 17.8979,
|
| 252 |
+
"mean_token_accuracy": 0.7612847778946161,
|
| 253 |
+
"num_tokens": 6622798.0,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.32,
|
| 258 |
+
"eval_biology_entropy": 2.7866374349594114,
|
| 259 |
+
"eval_biology_loss": 3.2049155235290527,
|
| 260 |
+
"eval_biology_mean_token_accuracy": 0.502942953824997,
|
| 261 |
+
"eval_biology_num_tokens": 6622798.0,
|
| 262 |
+
"eval_biology_runtime": 56.7587,
|
| 263 |
+
"eval_biology_samples_per_second": 8.809,
|
| 264 |
+
"eval_biology_steps_per_second": 2.202,
|
| 265 |
+
"step": 200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.32,
|
| 269 |
+
"eval_chemistry_entropy": 1.41602649641037,
|
| 270 |
+
"eval_chemistry_loss": 1.4916778802871704,
|
| 271 |
+
"eval_chemistry_mean_token_accuracy": 0.7079435839653015,
|
| 272 |
+
"eval_chemistry_num_tokens": 6622798.0,
|
| 273 |
+
"eval_chemistry_runtime": 71.0228,
|
| 274 |
+
"eval_chemistry_samples_per_second": 7.04,
|
| 275 |
+
"eval_chemistry_steps_per_second": 1.76,
|
| 276 |
+
"step": 200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.32,
|
| 280 |
+
"eval_math_entropy": 0.9821910600662231,
|
| 281 |
+
"eval_math_loss": 1.256189227104187,
|
| 282 |
+
"eval_math_mean_token_accuracy": 0.7425830450057983,
|
| 283 |
+
"eval_math_num_tokens": 6622798.0,
|
| 284 |
+
"eval_math_runtime": 75.4477,
|
| 285 |
+
"eval_math_samples_per_second": 6.627,
|
| 286 |
+
"eval_math_steps_per_second": 1.657,
|
| 287 |
+
"step": 200
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.32,
|
| 291 |
+
"eval_physics_entropy": 1.0466554942131043,
|
| 292 |
+
"eval_physics_loss": 1.1129223108291626,
|
| 293 |
+
"eval_physics_mean_token_accuracy": 0.7617223796844482,
|
| 294 |
+
"eval_physics_num_tokens": 6622798.0,
|
| 295 |
+
"eval_physics_runtime": 85.3386,
|
| 296 |
+
"eval_physics_samples_per_second": 5.859,
|
| 297 |
+
"eval_physics_steps_per_second": 1.465,
|
| 298 |
+
"step": 200
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"entropy": 1.0462193094193935,
|
| 302 |
+
"epoch": 0.336,
|
| 303 |
+
"grad_norm": 11.875,
|
| 304 |
+
"learning_rate": 1.3933333333333334e-05,
|
| 305 |
+
"loss": 17.7296,
|
| 306 |
+
"mean_token_accuracy": 0.7623197592794895,
|
| 307 |
+
"num_tokens": 6953611.0,
|
| 308 |
+
"step": 210
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"entropy": 1.0164767142385245,
|
| 312 |
+
"epoch": 0.352,
|
| 313 |
+
"grad_norm": 11.25,
|
| 314 |
+
"learning_rate": 1.46e-05,
|
| 315 |
+
"loss": 17.1525,
|
| 316 |
+
"mean_token_accuracy": 0.7695928748697043,
|
| 317 |
+
"num_tokens": 7290378.0,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"entropy": 0.9987690806388855,
|
| 322 |
+
"epoch": 0.368,
|
| 323 |
+
"grad_norm": 11.4375,
|
| 324 |
+
"learning_rate": 1.5266666666666667e-05,
|
| 325 |
+
"loss": 16.9228,
|
| 326 |
+
"mean_token_accuracy": 0.7714470315724611,
|
| 327 |
+
"num_tokens": 7621082.0,
|
| 328 |
+
"step": 230
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"entropy": 1.0506686847656965,
|
| 332 |
+
"epoch": 0.384,
|
| 333 |
+
"grad_norm": 11.3125,
|
| 334 |
+
"learning_rate": 1.5933333333333336e-05,
|
| 335 |
+
"loss": 17.6416,
|
| 336 |
+
"mean_token_accuracy": 0.7599896423518657,
|
| 337 |
+
"num_tokens": 7955570.0,
|
| 338 |
+
"step": 240
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"entropy": 1.0135670874267817,
|
| 342 |
+
"epoch": 0.4,
|
| 343 |
+
"grad_norm": 14.5,
|
| 344 |
+
"learning_rate": 1.66e-05,
|
| 345 |
+
"loss": 16.9698,
|
| 346 |
+
"mean_token_accuracy": 0.7687441952526569,
|
| 347 |
+
"num_tokens": 8291049.0,
|
| 348 |
+
"step": 250
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"entropy": 0.9876859273761511,
|
| 352 |
+
"epoch": 0.416,
|
| 353 |
+
"grad_norm": 13.75,
|
| 354 |
+
"learning_rate": 1.726666666666667e-05,
|
| 355 |
+
"loss": 16.6909,
|
| 356 |
+
"mean_token_accuracy": 0.7725014701485634,
|
| 357 |
+
"num_tokens": 8627310.0,
|
| 358 |
+
"step": 260
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"entropy": 1.0216444052755833,
|
| 362 |
+
"epoch": 0.432,
|
| 363 |
+
"grad_norm": 12.125,
|
| 364 |
+
"learning_rate": 1.7933333333333333e-05,
|
| 365 |
+
"loss": 17.011,
|
| 366 |
+
"mean_token_accuracy": 0.7689431305974722,
|
| 367 |
+
"num_tokens": 8958371.0,
|
| 368 |
+
"step": 270
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"entropy": 0.9649551097303629,
|
| 372 |
+
"epoch": 0.448,
|
| 373 |
+
"grad_norm": 12.1875,
|
| 374 |
+
"learning_rate": 1.86e-05,
|
| 375 |
+
"loss": 16.2252,
|
| 376 |
+
"mean_token_accuracy": 0.7780237648636102,
|
| 377 |
+
"num_tokens": 9294388.0,
|
| 378 |
+
"step": 280
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"entropy": 0.9653072291985154,
|
| 382 |
+
"epoch": 0.464,
|
| 383 |
+
"grad_norm": 12.5,
|
| 384 |
+
"learning_rate": 1.926666666666667e-05,
|
| 385 |
+
"loss": 16.1144,
|
| 386 |
+
"mean_token_accuracy": 0.7772431872785092,
|
| 387 |
+
"num_tokens": 9620366.0,
|
| 388 |
+
"step": 290
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"entropy": 0.9695413928478956,
|
| 392 |
+
"epoch": 0.48,
|
| 393 |
+
"grad_norm": 11.8125,
|
| 394 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 395 |
+
"loss": 16.4447,
|
| 396 |
+
"mean_token_accuracy": 0.7733453687280416,
|
| 397 |
+
"num_tokens": 9955431.0,
|
| 398 |
+
"step": 300
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.48,
|
| 402 |
+
"eval_biology_entropy": 2.7462952222824097,
|
| 403 |
+
"eval_biology_loss": 3.060053825378418,
|
| 404 |
+
"eval_biology_mean_token_accuracy": 0.5114872539043427,
|
| 405 |
+
"eval_biology_num_tokens": 9955431.0,
|
| 406 |
+
"eval_biology_runtime": 56.9805,
|
| 407 |
+
"eval_biology_samples_per_second": 8.775,
|
| 408 |
+
"eval_biology_steps_per_second": 2.194,
|
| 409 |
+
"step": 300
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.48,
|
| 413 |
+
"eval_chemistry_entropy": 1.3710836601257324,
|
| 414 |
+
"eval_chemistry_loss": 1.4066872596740723,
|
| 415 |
+
"eval_chemistry_mean_token_accuracy": 0.7168692064285278,
|
| 416 |
+
"eval_chemistry_num_tokens": 9955431.0,
|
| 417 |
+
"eval_chemistry_runtime": 71.3717,
|
| 418 |
+
"eval_chemistry_samples_per_second": 7.006,
|
| 419 |
+
"eval_chemistry_steps_per_second": 1.751,
|
| 420 |
+
"step": 300
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.48,
|
| 424 |
+
"eval_math_entropy": 0.9784291224479675,
|
| 425 |
+
"eval_math_loss": 1.2082456350326538,
|
| 426 |
+
"eval_math_mean_token_accuracy": 0.7478420066833497,
|
| 427 |
+
"eval_math_num_tokens": 9955431.0,
|
| 428 |
+
"eval_math_runtime": 74.3706,
|
| 429 |
+
"eval_math_samples_per_second": 6.723,
|
| 430 |
+
"eval_math_steps_per_second": 1.681,
|
| 431 |
+
"step": 300
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.48,
|
| 435 |
+
"eval_physics_entropy": 1.007008915424347,
|
| 436 |
+
"eval_physics_loss": 1.0401662588119507,
|
| 437 |
+
"eval_physics_mean_token_accuracy": 0.7707204632759094,
|
| 438 |
+
"eval_physics_num_tokens": 9955431.0,
|
| 439 |
+
"eval_physics_runtime": 84.6939,
|
| 440 |
+
"eval_physics_samples_per_second": 5.904,
|
| 441 |
+
"eval_physics_steps_per_second": 1.476,
|
| 442 |
+
"step": 300
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"entropy": 1.007104156538844,
|
| 446 |
+
"epoch": 0.496,
|
| 447 |
+
"grad_norm": 10.5,
|
| 448 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 449 |
+
"loss": 16.6639,
|
| 450 |
+
"mean_token_accuracy": 0.771253039687872,
|
| 451 |
+
"num_tokens": 10289883.0,
|
| 452 |
+
"step": 310
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"entropy": 0.9723599992692471,
|
| 456 |
+
"epoch": 0.512,
|
| 457 |
+
"grad_norm": 12.4375,
|
| 458 |
+
"learning_rate": 1.985925925925926e-05,
|
| 459 |
+
"loss": 16.4387,
|
| 460 |
+
"mean_token_accuracy": 0.7748051699250936,
|
| 461 |
+
"num_tokens": 10619468.0,
|
| 462 |
+
"step": 320
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"entropy": 0.9979474730789661,
|
| 466 |
+
"epoch": 0.528,
|
| 467 |
+
"grad_norm": 10.8125,
|
| 468 |
+
"learning_rate": 1.9785185185185187e-05,
|
| 469 |
+
"loss": 16.6398,
|
| 470 |
+
"mean_token_accuracy": 0.7712166555225849,
|
| 471 |
+
"num_tokens": 10944904.0,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"entropy": 0.9800913408398628,
|
| 476 |
+
"epoch": 0.544,
|
| 477 |
+
"grad_norm": 11.625,
|
| 478 |
+
"learning_rate": 1.971111111111111e-05,
|
| 479 |
+
"loss": 16.374,
|
| 480 |
+
"mean_token_accuracy": 0.7737775973975658,
|
| 481 |
+
"num_tokens": 11266409.0,
|
| 482 |
+
"step": 340
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"entropy": 0.9599094491451978,
|
| 486 |
+
"epoch": 0.56,
|
| 487 |
+
"grad_norm": 10.625,
|
| 488 |
+
"learning_rate": 1.963703703703704e-05,
|
| 489 |
+
"loss": 16.0194,
|
| 490 |
+
"mean_token_accuracy": 0.778597004711628,
|
| 491 |
+
"num_tokens": 11605544.0,
|
| 492 |
+
"step": 350
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"entropy": 0.948445113003254,
|
| 496 |
+
"epoch": 0.576,
|
| 497 |
+
"grad_norm": 11.75,
|
| 498 |
+
"learning_rate": 1.9562962962962964e-05,
|
| 499 |
+
"loss": 15.9791,
|
| 500 |
+
"mean_token_accuracy": 0.7805619373917579,
|
| 501 |
+
"num_tokens": 11931715.0,
|
| 502 |
+
"step": 360
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"entropy": 0.9618662863969802,
|
| 506 |
+
"epoch": 0.592,
|
| 507 |
+
"grad_norm": 13.0625,
|
| 508 |
+
"learning_rate": 1.948888888888889e-05,
|
| 509 |
+
"loss": 16.0713,
|
| 510 |
+
"mean_token_accuracy": 0.776887471601367,
|
| 511 |
+
"num_tokens": 12257243.0,
|
| 512 |
+
"step": 370
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"entropy": 0.9480277199298144,
|
| 516 |
+
"epoch": 0.608,
|
| 517 |
+
"grad_norm": 10.8125,
|
| 518 |
+
"learning_rate": 1.9414814814814817e-05,
|
| 519 |
+
"loss": 15.8656,
|
| 520 |
+
"mean_token_accuracy": 0.7797395702451467,
|
| 521 |
+
"num_tokens": 12583154.0,
|
| 522 |
+
"step": 380
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"entropy": 0.9716805059462785,
|
| 526 |
+
"epoch": 0.624,
|
| 527 |
+
"grad_norm": 10.625,
|
| 528 |
+
"learning_rate": 1.9340740740740743e-05,
|
| 529 |
+
"loss": 16.1886,
|
| 530 |
+
"mean_token_accuracy": 0.7762768242508173,
|
| 531 |
+
"num_tokens": 12905392.0,
|
| 532 |
+
"step": 390
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"entropy": 0.9691856369376183,
|
| 536 |
+
"epoch": 0.64,
|
| 537 |
+
"grad_norm": 11.8125,
|
| 538 |
+
"learning_rate": 1.926666666666667e-05,
|
| 539 |
+
"loss": 16.2314,
|
| 540 |
+
"mean_token_accuracy": 0.7761619180440903,
|
| 541 |
+
"num_tokens": 13232198.0,
|
| 542 |
+
"step": 400
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.64,
|
| 546 |
+
"eval_biology_entropy": 2.67273916721344,
|
| 547 |
+
"eval_biology_loss": 2.96358585357666,
|
| 548 |
+
"eval_biology_mean_token_accuracy": 0.5216054661273957,
|
| 549 |
+
"eval_biology_num_tokens": 13232198.0,
|
| 550 |
+
"eval_biology_runtime": 57.8851,
|
| 551 |
+
"eval_biology_samples_per_second": 8.638,
|
| 552 |
+
"eval_biology_steps_per_second": 2.159,
|
| 553 |
+
"step": 400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.64,
|
| 557 |
+
"eval_chemistry_entropy": 1.3173069462776184,
|
| 558 |
+
"eval_chemistry_loss": 1.36112380027771,
|
| 559 |
+
"eval_chemistry_mean_token_accuracy": 0.7232093935012818,
|
| 560 |
+
"eval_chemistry_num_tokens": 13232198.0,
|
| 561 |
+
"eval_chemistry_runtime": 71.8687,
|
| 562 |
+
"eval_chemistry_samples_per_second": 6.957,
|
| 563 |
+
"eval_chemistry_steps_per_second": 1.739,
|
| 564 |
+
"step": 400
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.64,
|
| 568 |
+
"eval_math_entropy": 0.9528153438568115,
|
| 569 |
+
"eval_math_loss": 1.1841319799423218,
|
| 570 |
+
"eval_math_mean_token_accuracy": 0.7507999067306519,
|
| 571 |
+
"eval_math_num_tokens": 13232198.0,
|
| 572 |
+
"eval_math_runtime": 74.8756,
|
| 573 |
+
"eval_math_samples_per_second": 6.678,
|
| 574 |
+
"eval_math_steps_per_second": 1.669,
|
| 575 |
+
"step": 400
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.64,
|
| 579 |
+
"eval_physics_entropy": 0.9614342203140259,
|
| 580 |
+
"eval_physics_loss": 1.0016775131225586,
|
| 581 |
+
"eval_physics_mean_token_accuracy": 0.7765112986564636,
|
| 582 |
+
"eval_physics_num_tokens": 13232198.0,
|
| 583 |
+
"eval_physics_runtime": 85.1175,
|
| 584 |
+
"eval_physics_samples_per_second": 5.874,
|
| 585 |
+
"eval_physics_steps_per_second": 1.469,
|
| 586 |
+
"step": 400
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"entropy": 0.9530647125095129,
|
| 590 |
+
"epoch": 0.656,
|
| 591 |
+
"grad_norm": 11.375,
|
| 592 |
+
"learning_rate": 1.9192592592592593e-05,
|
| 593 |
+
"loss": 15.9429,
|
| 594 |
+
"mean_token_accuracy": 0.7780249908566474,
|
| 595 |
+
"num_tokens": 13575902.0,
|
| 596 |
+
"step": 410
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"entropy": 0.9399624351412058,
|
| 600 |
+
"epoch": 0.672,
|
| 601 |
+
"grad_norm": 12.0625,
|
| 602 |
+
"learning_rate": 1.911851851851852e-05,
|
| 603 |
+
"loss": 15.7284,
|
| 604 |
+
"mean_token_accuracy": 0.7817466016858816,
|
| 605 |
+
"num_tokens": 13895997.0,
|
| 606 |
+
"step": 420
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"entropy": 0.9540142439305782,
|
| 610 |
+
"epoch": 0.688,
|
| 611 |
+
"grad_norm": 10.6875,
|
| 612 |
+
"learning_rate": 1.9044444444444446e-05,
|
| 613 |
+
"loss": 15.7763,
|
| 614 |
+
"mean_token_accuracy": 0.7817316662520171,
|
| 615 |
+
"num_tokens": 14234888.0,
|
| 616 |
+
"step": 430
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"entropy": 0.9538515221327544,
|
| 620 |
+
"epoch": 0.704,
|
| 621 |
+
"grad_norm": 11.625,
|
| 622 |
+
"learning_rate": 1.8970370370370372e-05,
|
| 623 |
+
"loss": 16.0838,
|
| 624 |
+
"mean_token_accuracy": 0.775813952833414,
|
| 625 |
+
"num_tokens": 14567908.0,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"entropy": 0.9464580094441771,
|
| 630 |
+
"epoch": 0.72,
|
| 631 |
+
"grad_norm": 12.375,
|
| 632 |
+
"learning_rate": 1.8896296296296295e-05,
|
| 633 |
+
"loss": 15.704,
|
| 634 |
+
"mean_token_accuracy": 0.7821726374328136,
|
| 635 |
+
"num_tokens": 14882927.0,
|
| 636 |
+
"step": 450
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"entropy": 0.9363573594018817,
|
| 640 |
+
"epoch": 0.736,
|
| 641 |
+
"grad_norm": 13.0625,
|
| 642 |
+
"learning_rate": 1.8822222222222225e-05,
|
| 643 |
+
"loss": 15.7075,
|
| 644 |
+
"mean_token_accuracy": 0.7809599131345749,
|
| 645 |
+
"num_tokens": 15217342.0,
|
| 646 |
+
"step": 460
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"entropy": 0.9193624388426542,
|
| 650 |
+
"epoch": 0.752,
|
| 651 |
+
"grad_norm": 10.4375,
|
| 652 |
+
"learning_rate": 1.874814814814815e-05,
|
| 653 |
+
"loss": 15.3677,
|
| 654 |
+
"mean_token_accuracy": 0.7846659943461418,
|
| 655 |
+
"num_tokens": 15547710.0,
|
| 656 |
+
"step": 470
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"entropy": 0.9613647162914276,
|
| 660 |
+
"epoch": 0.768,
|
| 661 |
+
"grad_norm": 11.1875,
|
| 662 |
+
"learning_rate": 1.8674074074074075e-05,
|
| 663 |
+
"loss": 16.0934,
|
| 664 |
+
"mean_token_accuracy": 0.7759233452379704,
|
| 665 |
+
"num_tokens": 15877177.0,
|
| 666 |
+
"step": 480
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"entropy": 0.9139260027557612,
|
| 670 |
+
"epoch": 0.784,
|
| 671 |
+
"grad_norm": 10.1875,
|
| 672 |
+
"learning_rate": 1.86e-05,
|
| 673 |
+
"loss": 15.3216,
|
| 674 |
+
"mean_token_accuracy": 0.786144433543086,
|
| 675 |
+
"num_tokens": 16219640.0,
|
| 676 |
+
"step": 490
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"entropy": 0.9349881060421467,
|
| 680 |
+
"epoch": 0.8,
|
| 681 |
+
"grad_norm": 12.0,
|
| 682 |
+
"learning_rate": 1.8525925925925928e-05,
|
| 683 |
+
"loss": 15.6534,
|
| 684 |
+
"mean_token_accuracy": 0.7819662269204855,
|
| 685 |
+
"num_tokens": 16548261.0,
|
| 686 |
+
"step": 500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.8,
|
| 690 |
+
"eval_biology_entropy": 2.617759398460388,
|
| 691 |
+
"eval_biology_loss": 2.9445557594299316,
|
| 692 |
+
"eval_biology_mean_token_accuracy": 0.5238966343402862,
|
| 693 |
+
"eval_biology_num_tokens": 16548261.0,
|
| 694 |
+
"eval_biology_runtime": 57.6905,
|
| 695 |
+
"eval_biology_samples_per_second": 8.667,
|
| 696 |
+
"eval_biology_steps_per_second": 2.167,
|
| 697 |
+
"step": 500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.8,
|
| 701 |
+
"eval_chemistry_entropy": 1.2937732858657838,
|
| 702 |
+
"eval_chemistry_loss": 1.3414490222930908,
|
| 703 |
+
"eval_chemistry_mean_token_accuracy": 0.7248484263420105,
|
| 704 |
+
"eval_chemistry_num_tokens": 16548261.0,
|
| 705 |
+
"eval_chemistry_runtime": 71.1659,
|
| 706 |
+
"eval_chemistry_samples_per_second": 7.026,
|
| 707 |
+
"eval_chemistry_steps_per_second": 1.756,
|
| 708 |
+
"step": 500
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.8,
|
| 712 |
+
"eval_math_entropy": 0.93889684009552,
|
| 713 |
+
"eval_math_loss": 1.17385995388031,
|
| 714 |
+
"eval_math_mean_token_accuracy": 0.7522515163421631,
|
| 715 |
+
"eval_math_num_tokens": 16548261.0,
|
| 716 |
+
"eval_math_runtime": 74.7142,
|
| 717 |
+
"eval_math_samples_per_second": 6.692,
|
| 718 |
+
"eval_math_steps_per_second": 1.673,
|
| 719 |
+
"step": 500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.8,
|
| 723 |
+
"eval_physics_entropy": 0.9420124840736389,
|
| 724 |
+
"eval_physics_loss": 0.9809063673019409,
|
| 725 |
+
"eval_physics_mean_token_accuracy": 0.7797514395713806,
|
| 726 |
+
"eval_physics_num_tokens": 16548261.0,
|
| 727 |
+
"eval_physics_runtime": 84.5204,
|
| 728 |
+
"eval_physics_samples_per_second": 5.916,
|
| 729 |
+
"eval_physics_steps_per_second": 1.479,
|
| 730 |
+
"step": 500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 0.9288215478882194,
|
| 734 |
+
"epoch": 0.816,
|
| 735 |
+
"grad_norm": 11.3125,
|
| 736 |
+
"learning_rate": 1.8451851851851855e-05,
|
| 737 |
+
"loss": 15.5725,
|
| 738 |
+
"mean_token_accuracy": 0.7813547737896442,
|
| 739 |
+
"num_tokens": 16870533.0,
|
| 740 |
+
"step": 510
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 0.9446330968290567,
|
| 744 |
+
"epoch": 0.832,
|
| 745 |
+
"grad_norm": 15.875,
|
| 746 |
+
"learning_rate": 1.8377777777777778e-05,
|
| 747 |
+
"loss": 15.7952,
|
| 748 |
+
"mean_token_accuracy": 0.7800830528140068,
|
| 749 |
+
"num_tokens": 17196371.0,
|
| 750 |
+
"step": 520
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 0.9318377874791622,
|
| 754 |
+
"epoch": 0.848,
|
| 755 |
+
"grad_norm": 10.625,
|
| 756 |
+
"learning_rate": 1.8303703703703704e-05,
|
| 757 |
+
"loss": 15.4484,
|
| 758 |
+
"mean_token_accuracy": 0.7836398217827082,
|
| 759 |
+
"num_tokens": 17527066.0,
|
| 760 |
+
"step": 530
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 0.9298348639160394,
|
| 764 |
+
"epoch": 0.864,
|
| 765 |
+
"grad_norm": 10.625,
|
| 766 |
+
"learning_rate": 1.822962962962963e-05,
|
| 767 |
+
"loss": 15.6026,
|
| 768 |
+
"mean_token_accuracy": 0.7814645994454622,
|
| 769 |
+
"num_tokens": 17859605.0,
|
| 770 |
+
"step": 540
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 0.9549105998128653,
|
| 774 |
+
"epoch": 0.88,
|
| 775 |
+
"grad_norm": 12.625,
|
| 776 |
+
"learning_rate": 1.8155555555555557e-05,
|
| 777 |
+
"loss": 15.7997,
|
| 778 |
+
"mean_token_accuracy": 0.7805559232831001,
|
| 779 |
+
"num_tokens": 18187315.0,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 0.8820257507264614,
|
| 784 |
+
"epoch": 0.896,
|
| 785 |
+
"grad_norm": 10.5625,
|
| 786 |
+
"learning_rate": 1.8081481481481484e-05,
|
| 787 |
+
"loss": 14.7783,
|
| 788 |
+
"mean_token_accuracy": 0.7907331828027964,
|
| 789 |
+
"num_tokens": 18536790.0,
|
| 790 |
+
"step": 560
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 0.9362711973488331,
|
| 794 |
+
"epoch": 0.912,
|
| 795 |
+
"grad_norm": 12.5625,
|
| 796 |
+
"learning_rate": 1.800740740740741e-05,
|
| 797 |
+
"loss": 15.6226,
|
| 798 |
+
"mean_token_accuracy": 0.7802051860839129,
|
| 799 |
+
"num_tokens": 18869344.0,
|
| 800 |
+
"step": 570
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.9295620009303093,
|
| 804 |
+
"epoch": 0.928,
|
| 805 |
+
"grad_norm": 12.75,
|
| 806 |
+
"learning_rate": 1.7933333333333333e-05,
|
| 807 |
+
"loss": 15.5339,
|
| 808 |
+
"mean_token_accuracy": 0.7815026968717576,
|
| 809 |
+
"num_tokens": 19199834.0,
|
| 810 |
+
"step": 580
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 0.9112746389582753,
|
| 814 |
+
"epoch": 0.944,
|
| 815 |
+
"grad_norm": 12.5,
|
| 816 |
+
"learning_rate": 1.785925925925926e-05,
|
| 817 |
+
"loss": 15.2153,
|
| 818 |
+
"mean_token_accuracy": 0.7855059750378132,
|
| 819 |
+
"num_tokens": 19525385.0,
|
| 820 |
+
"step": 590
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 0.9181267462670804,
|
| 824 |
+
"epoch": 0.96,
|
| 825 |
+
"grad_norm": 12.75,
|
| 826 |
+
"learning_rate": 1.7785185185185186e-05,
|
| 827 |
+
"loss": 15.3751,
|
| 828 |
+
"mean_token_accuracy": 0.7829241104424,
|
| 829 |
+
"num_tokens": 19856973.0,
|
| 830 |
+
"step": 600
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.96,
|
| 834 |
+
"eval_biology_entropy": 2.594178828239441,
|
| 835 |
+
"eval_biology_loss": 2.9207751750946045,
|
| 836 |
+
"eval_biology_mean_token_accuracy": 0.5271391036510468,
|
| 837 |
+
"eval_biology_num_tokens": 19856973.0,
|
| 838 |
+
"eval_biology_runtime": 57.0207,
|
| 839 |
+
"eval_biology_samples_per_second": 8.769,
|
| 840 |
+
"eval_biology_steps_per_second": 2.192,
|
| 841 |
+
"step": 600
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.96,
|
| 845 |
+
"eval_chemistry_entropy": 1.270768217086792,
|
| 846 |
+
"eval_chemistry_loss": 1.3242253065109253,
|
| 847 |
+
"eval_chemistry_mean_token_accuracy": 0.7280859136581421,
|
| 848 |
+
"eval_chemistry_num_tokens": 19856973.0,
|
| 849 |
+
"eval_chemistry_runtime": 72.006,
|
| 850 |
+
"eval_chemistry_samples_per_second": 6.944,
|
| 851 |
+
"eval_chemistry_steps_per_second": 1.736,
|
| 852 |
+
"step": 600
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.96,
|
| 856 |
+
"eval_math_entropy": 0.9325860266685486,
|
| 857 |
+
"eval_math_loss": 1.1655522584915161,
|
| 858 |
+
"eval_math_mean_token_accuracy": 0.7533689908981324,
|
| 859 |
+
"eval_math_num_tokens": 19856973.0,
|
| 860 |
+
"eval_math_runtime": 74.0585,
|
| 861 |
+
"eval_math_samples_per_second": 6.751,
|
| 862 |
+
"eval_math_steps_per_second": 1.688,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.96,
|
| 867 |
+
"eval_physics_entropy": 0.9269161529541016,
|
| 868 |
+
"eval_physics_loss": 0.9667912125587463,
|
| 869 |
+
"eval_physics_mean_token_accuracy": 0.7814396271705627,
|
| 870 |
+
"eval_physics_num_tokens": 19856973.0,
|
| 871 |
+
"eval_physics_runtime": 84.5863,
|
| 872 |
+
"eval_physics_samples_per_second": 5.911,
|
| 873 |
+
"eval_physics_steps_per_second": 1.478,
|
| 874 |
+
"step": 600
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"entropy": 0.9162795972079039,
|
| 878 |
+
"epoch": 0.976,
|
| 879 |
+
"grad_norm": 12.125,
|
| 880 |
+
"learning_rate": 1.7711111111111113e-05,
|
| 881 |
+
"loss": 15.2813,
|
| 882 |
+
"mean_token_accuracy": 0.7860829800367355,
|
| 883 |
+
"num_tokens": 20190600.0,
|
| 884 |
+
"step": 610
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"entropy": 0.9235894799232482,
|
| 888 |
+
"epoch": 0.992,
|
| 889 |
+
"grad_norm": 11.5,
|
| 890 |
+
"learning_rate": 1.763703703703704e-05,
|
| 891 |
+
"loss": 15.4241,
|
| 892 |
+
"mean_token_accuracy": 0.784028148651123,
|
| 893 |
+
"num_tokens": 20518014.0,
|
| 894 |
+
"step": 620
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"entropy": 0.9041305113583803,
|
| 898 |
+
"epoch": 1.008,
|
| 899 |
+
"grad_norm": 11.0625,
|
| 900 |
+
"learning_rate": 1.7562962962962962e-05,
|
| 901 |
+
"loss": 14.9099,
|
| 902 |
+
"mean_token_accuracy": 0.7884359251707792,
|
| 903 |
+
"num_tokens": 20859215.0,
|
| 904 |
+
"step": 630
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"entropy": 0.9018714495003224,
|
| 908 |
+
"epoch": 1.024,
|
| 909 |
+
"grad_norm": 11.1875,
|
| 910 |
+
"learning_rate": 1.7488888888888892e-05,
|
| 911 |
+
"loss": 15.1658,
|
| 912 |
+
"mean_token_accuracy": 0.7857498530298471,
|
| 913 |
+
"num_tokens": 21186658.0,
|
| 914 |
+
"step": 640
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"entropy": 0.916073745302856,
|
| 918 |
+
"epoch": 1.04,
|
| 919 |
+
"grad_norm": 10.8125,
|
| 920 |
+
"learning_rate": 1.7414814814814815e-05,
|
| 921 |
+
"loss": 15.1367,
|
| 922 |
+
"mean_token_accuracy": 0.7867844242602586,
|
| 923 |
+
"num_tokens": 21517237.0,
|
| 924 |
+
"step": 650
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"entropy": 0.8802109774202108,
|
| 928 |
+
"epoch": 1.056,
|
| 929 |
+
"grad_norm": 11.0,
|
| 930 |
+
"learning_rate": 1.7340740740740742e-05,
|
| 931 |
+
"loss": 14.7961,
|
| 932 |
+
"mean_token_accuracy": 0.7896511305123568,
|
| 933 |
+
"num_tokens": 21852528.0,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"entropy": 0.9102928217500448,
|
| 938 |
+
"epoch": 1.072,
|
| 939 |
+
"grad_norm": 10.0,
|
| 940 |
+
"learning_rate": 1.726666666666667e-05,
|
| 941 |
+
"loss": 15.1014,
|
| 942 |
+
"mean_token_accuracy": 0.7862772591412067,
|
| 943 |
+
"num_tokens": 22192380.0,
|
| 944 |
+
"step": 670
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"entropy": 0.9107371259480714,
|
| 948 |
+
"epoch": 1.088,
|
| 949 |
+
"grad_norm": 13.4375,
|
| 950 |
+
"learning_rate": 1.7192592592592595e-05,
|
| 951 |
+
"loss": 15.2493,
|
| 952 |
+
"mean_token_accuracy": 0.7848471954464913,
|
| 953 |
+
"num_tokens": 22517971.0,
|
| 954 |
+
"step": 680
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"entropy": 0.8938208010047675,
|
| 958 |
+
"epoch": 1.104,
|
| 959 |
+
"grad_norm": 10.6875,
|
| 960 |
+
"learning_rate": 1.711851851851852e-05,
|
| 961 |
+
"loss": 14.9017,
|
| 962 |
+
"mean_token_accuracy": 0.7887144055217504,
|
| 963 |
+
"num_tokens": 22848694.0,
|
| 964 |
+
"step": 690
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"entropy": 0.9129897281527519,
|
| 968 |
+
"epoch": 1.12,
|
| 969 |
+
"grad_norm": 11.4375,
|
| 970 |
+
"learning_rate": 1.7044444444444445e-05,
|
| 971 |
+
"loss": 15.2722,
|
| 972 |
+
"mean_token_accuracy": 0.7840389590710402,
|
| 973 |
+
"num_tokens": 23183889.0,
|
| 974 |
+
"step": 700
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.12,
|
| 978 |
+
"eval_biology_entropy": 2.5242818908691405,
|
| 979 |
+
"eval_biology_loss": 2.9092111587524414,
|
| 980 |
+
"eval_biology_mean_token_accuracy": 0.5278490686416626,
|
| 981 |
+
"eval_biology_num_tokens": 23183889.0,
|
| 982 |
+
"eval_biology_runtime": 57.2117,
|
| 983 |
+
"eval_biology_samples_per_second": 8.739,
|
| 984 |
+
"eval_biology_steps_per_second": 2.185,
|
| 985 |
+
"step": 700
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 1.12,
|
| 989 |
+
"eval_chemistry_entropy": 1.2367294158935547,
|
| 990 |
+
"eval_chemistry_loss": 1.3155030012130737,
|
| 991 |
+
"eval_chemistry_mean_token_accuracy": 0.7286214451789856,
|
| 992 |
+
"eval_chemistry_num_tokens": 23183889.0,
|
| 993 |
+
"eval_chemistry_runtime": 71.8224,
|
| 994 |
+
"eval_chemistry_samples_per_second": 6.962,
|
| 995 |
+
"eval_chemistry_steps_per_second": 1.74,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.12,
|
| 1000 |
+
"eval_math_entropy": 0.9127886519432068,
|
| 1001 |
+
"eval_math_loss": 1.1603002548217773,
|
| 1002 |
+
"eval_math_mean_token_accuracy": 0.754603410243988,
|
| 1003 |
+
"eval_math_num_tokens": 23183889.0,
|
| 1004 |
+
"eval_math_runtime": 74.5403,
|
| 1005 |
+
"eval_math_samples_per_second": 6.708,
|
| 1006 |
+
"eval_math_steps_per_second": 1.677,
|
| 1007 |
+
"step": 700
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 1.12,
|
| 1011 |
+
"eval_physics_entropy": 0.9048804240226745,
|
| 1012 |
+
"eval_physics_loss": 0.9561019539833069,
|
| 1013 |
+
"eval_physics_mean_token_accuracy": 0.7832872924804688,
|
| 1014 |
+
"eval_physics_num_tokens": 23183889.0,
|
| 1015 |
+
"eval_physics_runtime": 84.6107,
|
| 1016 |
+
"eval_physics_samples_per_second": 5.909,
|
| 1017 |
+
"eval_physics_steps_per_second": 1.477,
|
| 1018 |
+
"step": 700
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"entropy": 0.8741975981742144,
|
| 1022 |
+
"epoch": 1.1360000000000001,
|
| 1023 |
+
"grad_norm": 10.1875,
|
| 1024 |
+
"learning_rate": 1.697037037037037e-05,
|
| 1025 |
+
"loss": 14.564,
|
| 1026 |
+
"mean_token_accuracy": 0.7930445209145546,
|
| 1027 |
+
"num_tokens": 23523903.0,
|
| 1028 |
+
"step": 710
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"entropy": 0.8917050797492265,
|
| 1032 |
+
"epoch": 1.152,
|
| 1033 |
+
"grad_norm": 12.75,
|
| 1034 |
+
"learning_rate": 1.6896296296296298e-05,
|
| 1035 |
+
"loss": 14.9557,
|
| 1036 |
+
"mean_token_accuracy": 0.7896463591605425,
|
| 1037 |
+
"num_tokens": 23848848.0,
|
| 1038 |
+
"step": 720
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"entropy": 0.9238279532641173,
|
| 1042 |
+
"epoch": 1.168,
|
| 1043 |
+
"grad_norm": 11.6875,
|
| 1044 |
+
"learning_rate": 1.6822222222222224e-05,
|
| 1045 |
+
"loss": 15.2756,
|
| 1046 |
+
"mean_token_accuracy": 0.7821665916591882,
|
| 1047 |
+
"num_tokens": 24174589.0,
|
| 1048 |
+
"step": 730
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"entropy": 0.9052068604156375,
|
| 1052 |
+
"epoch": 1.184,
|
| 1053 |
+
"grad_norm": 13.5625,
|
| 1054 |
+
"learning_rate": 1.6748148148148147e-05,
|
| 1055 |
+
"loss": 15.0924,
|
| 1056 |
+
"mean_token_accuracy": 0.7870742384344339,
|
| 1057 |
+
"num_tokens": 24495981.0,
|
| 1058 |
+
"step": 740
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"entropy": 0.877324876561761,
|
| 1062 |
+
"epoch": 1.2,
|
| 1063 |
+
"grad_norm": 11.1875,
|
| 1064 |
+
"learning_rate": 1.6674074074074077e-05,
|
| 1065 |
+
"loss": 14.7224,
|
| 1066 |
+
"mean_token_accuracy": 0.7919886518269778,
|
| 1067 |
+
"num_tokens": 24826245.0,
|
| 1068 |
+
"step": 750
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"entropy": 0.9162866469472647,
|
| 1072 |
+
"epoch": 1.216,
|
| 1073 |
+
"grad_norm": 14.1875,
|
| 1074 |
+
"learning_rate": 1.66e-05,
|
| 1075 |
+
"loss": 15.1252,
|
| 1076 |
+
"mean_token_accuracy": 0.7850615303963423,
|
| 1077 |
+
"num_tokens": 25147775.0,
|
| 1078 |
+
"step": 760
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"entropy": 0.8838450819253921,
|
| 1082 |
+
"epoch": 1.232,
|
| 1083 |
+
"grad_norm": 12.1875,
|
| 1084 |
+
"learning_rate": 1.6525925925925927e-05,
|
| 1085 |
+
"loss": 14.8829,
|
| 1086 |
+
"mean_token_accuracy": 0.7893829133361578,
|
| 1087 |
+
"num_tokens": 25480033.0,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"entropy": 0.8952008984982968,
|
| 1092 |
+
"epoch": 1.248,
|
| 1093 |
+
"grad_norm": 12.875,
|
| 1094 |
+
"learning_rate": 1.6451851851851853e-05,
|
| 1095 |
+
"loss": 14.9213,
|
| 1096 |
+
"mean_token_accuracy": 0.7889137178659439,
|
| 1097 |
+
"num_tokens": 25811978.0,
|
| 1098 |
+
"step": 780
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"entropy": 0.8893922835588455,
|
| 1102 |
+
"epoch": 1.264,
|
| 1103 |
+
"grad_norm": 12.6875,
|
| 1104 |
+
"learning_rate": 1.637777777777778e-05,
|
| 1105 |
+
"loss": 14.7989,
|
| 1106 |
+
"mean_token_accuracy": 0.7897166382521391,
|
| 1107 |
+
"num_tokens": 26146299.0,
|
| 1108 |
+
"step": 790
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"entropy": 0.9172267891466618,
|
| 1112 |
+
"epoch": 1.28,
|
| 1113 |
+
"grad_norm": 11.75,
|
| 1114 |
+
"learning_rate": 1.6303703703703706e-05,
|
| 1115 |
+
"loss": 15.3289,
|
| 1116 |
+
"mean_token_accuracy": 0.782142074778676,
|
| 1117 |
+
"num_tokens": 26478207.0,
|
| 1118 |
+
"step": 800
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.28,
|
| 1122 |
+
"eval_biology_entropy": 2.5403532466888428,
|
| 1123 |
+
"eval_biology_loss": 2.9071850776672363,
|
| 1124 |
+
"eval_biology_mean_token_accuracy": 0.529075275182724,
|
| 1125 |
+
"eval_biology_num_tokens": 26478207.0,
|
| 1126 |
+
"eval_biology_runtime": 56.934,
|
| 1127 |
+
"eval_biology_samples_per_second": 8.782,
|
| 1128 |
+
"eval_biology_steps_per_second": 2.196,
|
| 1129 |
+
"step": 800
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.28,
|
| 1133 |
+
"eval_chemistry_entropy": 1.234187596321106,
|
| 1134 |
+
"eval_chemistry_loss": 1.306308388710022,
|
| 1135 |
+
"eval_chemistry_mean_token_accuracy": 0.7299280157089233,
|
| 1136 |
+
"eval_chemistry_num_tokens": 26478207.0,
|
| 1137 |
+
"eval_chemistry_runtime": 71.938,
|
| 1138 |
+
"eval_chemistry_samples_per_second": 6.95,
|
| 1139 |
+
"eval_chemistry_steps_per_second": 1.738,
|
| 1140 |
+
"step": 800
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.28,
|
| 1144 |
+
"eval_math_entropy": 0.9124487805366516,
|
| 1145 |
+
"eval_math_loss": 1.1551406383514404,
|
| 1146 |
+
"eval_math_mean_token_accuracy": 0.7556050124168396,
|
| 1147 |
+
"eval_math_num_tokens": 26478207.0,
|
| 1148 |
+
"eval_math_runtime": 75.1703,
|
| 1149 |
+
"eval_math_samples_per_second": 6.652,
|
| 1150 |
+
"eval_math_steps_per_second": 1.663,
|
| 1151 |
+
"step": 800
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.28,
|
| 1155 |
+
"eval_physics_entropy": 0.8998128218650818,
|
| 1156 |
+
"eval_physics_loss": 0.9489150047302246,
|
| 1157 |
+
"eval_physics_mean_token_accuracy": 0.7840672006607056,
|
| 1158 |
+
"eval_physics_num_tokens": 26478207.0,
|
| 1159 |
+
"eval_physics_runtime": 85.0147,
|
| 1160 |
+
"eval_physics_samples_per_second": 5.881,
|
| 1161 |
+
"eval_physics_steps_per_second": 1.47,
|
| 1162 |
+
"step": 800
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"entropy": 0.8760543465614319,
|
| 1166 |
+
"epoch": 1.296,
|
| 1167 |
+
"grad_norm": 12.8125,
|
| 1168 |
+
"learning_rate": 1.622962962962963e-05,
|
| 1169 |
+
"loss": 14.6611,
|
| 1170 |
+
"mean_token_accuracy": 0.7923007346689701,
|
| 1171 |
+
"num_tokens": 26806477.0,
|
| 1172 |
+
"step": 810
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"entropy": 0.917656009271741,
|
| 1176 |
+
"epoch": 1.312,
|
| 1177 |
+
"grad_norm": 13.1875,
|
| 1178 |
+
"learning_rate": 1.6155555555555556e-05,
|
| 1179 |
+
"loss": 15.2379,
|
| 1180 |
+
"mean_token_accuracy": 0.7841715186834335,
|
| 1181 |
+
"num_tokens": 27137528.0,
|
| 1182 |
+
"step": 820
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"entropy": 0.8967037841677665,
|
| 1186 |
+
"epoch": 1.328,
|
| 1187 |
+
"grad_norm": 13.25,
|
| 1188 |
+
"learning_rate": 1.6081481481481482e-05,
|
| 1189 |
+
"loss": 14.9513,
|
| 1190 |
+
"mean_token_accuracy": 0.7878007985651493,
|
| 1191 |
+
"num_tokens": 27465859.0,
|
| 1192 |
+
"step": 830
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"entropy": 0.901091538183391,
|
| 1196 |
+
"epoch": 1.3439999999999999,
|
| 1197 |
+
"grad_norm": 13.375,
|
| 1198 |
+
"learning_rate": 1.600740740740741e-05,
|
| 1199 |
+
"loss": 15.1044,
|
| 1200 |
+
"mean_token_accuracy": 0.7861052300781012,
|
| 1201 |
+
"num_tokens": 27795568.0,
|
| 1202 |
+
"step": 840
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"entropy": 0.9121165912598371,
|
| 1206 |
+
"epoch": 1.3599999999999999,
|
| 1207 |
+
"grad_norm": 11.625,
|
| 1208 |
+
"learning_rate": 1.5933333333333336e-05,
|
| 1209 |
+
"loss": 15.0779,
|
| 1210 |
+
"mean_token_accuracy": 0.7859285034239292,
|
| 1211 |
+
"num_tokens": 28129988.0,
|
| 1212 |
+
"step": 850
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"entropy": 0.8819791020825505,
|
| 1216 |
+
"epoch": 1.376,
|
| 1217 |
+
"grad_norm": 13.875,
|
| 1218 |
+
"learning_rate": 1.5859259259259262e-05,
|
| 1219 |
+
"loss": 14.7997,
|
| 1220 |
+
"mean_token_accuracy": 0.7886631991714239,
|
| 1221 |
+
"num_tokens": 28465903.0,
|
| 1222 |
+
"step": 860
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"entropy": 0.8854727063328027,
|
| 1226 |
+
"epoch": 1.392,
|
| 1227 |
+
"grad_norm": 13.25,
|
| 1228 |
+
"learning_rate": 1.5785185185185185e-05,
|
| 1229 |
+
"loss": 14.61,
|
| 1230 |
+
"mean_token_accuracy": 0.7927087835967541,
|
| 1231 |
+
"num_tokens": 28801554.0,
|
| 1232 |
+
"step": 870
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"entropy": 0.8729526123031974,
|
| 1236 |
+
"epoch": 1.408,
|
| 1237 |
+
"grad_norm": 12.4375,
|
| 1238 |
+
"learning_rate": 1.571111111111111e-05,
|
| 1239 |
+
"loss": 14.7562,
|
| 1240 |
+
"mean_token_accuracy": 0.7899845536798239,
|
| 1241 |
+
"num_tokens": 29125114.0,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"entropy": 0.913954821228981,
|
| 1246 |
+
"epoch": 1.424,
|
| 1247 |
+
"grad_norm": 10.9375,
|
| 1248 |
+
"learning_rate": 1.5637037037037038e-05,
|
| 1249 |
+
"loss": 15.2188,
|
| 1250 |
+
"mean_token_accuracy": 0.783445942774415,
|
| 1251 |
+
"num_tokens": 29454334.0,
|
| 1252 |
+
"step": 890
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"entropy": 0.8652864851057529,
|
| 1256 |
+
"epoch": 1.44,
|
| 1257 |
+
"grad_norm": 12.1875,
|
| 1258 |
+
"learning_rate": 1.5562962962962965e-05,
|
| 1259 |
+
"loss": 14.3528,
|
| 1260 |
+
"mean_token_accuracy": 0.794481047987938,
|
| 1261 |
+
"num_tokens": 29789253.0,
|
| 1262 |
+
"step": 900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.44,
|
| 1266 |
+
"eval_biology_entropy": 2.4760996789932253,
|
| 1267 |
+
"eval_biology_loss": 2.897955894470215,
|
| 1268 |
+
"eval_biology_mean_token_accuracy": 0.5296812095642089,
|
| 1269 |
+
"eval_biology_num_tokens": 29789253.0,
|
| 1270 |
+
"eval_biology_runtime": 57.0842,
|
| 1271 |
+
"eval_biology_samples_per_second": 8.759,
|
| 1272 |
+
"eval_biology_steps_per_second": 2.19,
|
| 1273 |
+
"step": 900
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 1.44,
|
| 1277 |
+
"eval_chemistry_entropy": 1.2120891718864442,
|
| 1278 |
+
"eval_chemistry_loss": 1.2991372346878052,
|
| 1279 |
+
"eval_chemistry_mean_token_accuracy": 0.7307901701927185,
|
| 1280 |
+
"eval_chemistry_num_tokens": 29789253.0,
|
| 1281 |
+
"eval_chemistry_runtime": 71.5779,
|
| 1282 |
+
"eval_chemistry_samples_per_second": 6.985,
|
| 1283 |
+
"eval_chemistry_steps_per_second": 1.746,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.44,
|
| 1288 |
+
"eval_math_entropy": 0.9021294908523559,
|
| 1289 |
+
"eval_math_loss": 1.1532059907913208,
|
| 1290 |
+
"eval_math_mean_token_accuracy": 0.755294855594635,
|
| 1291 |
+
"eval_math_num_tokens": 29789253.0,
|
| 1292 |
+
"eval_math_runtime": 74.8641,
|
| 1293 |
+
"eval_math_samples_per_second": 6.679,
|
| 1294 |
+
"eval_math_steps_per_second": 1.67,
|
| 1295 |
+
"step": 900
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 1.44,
|
| 1299 |
+
"eval_physics_entropy": 0.8853664617538453,
|
| 1300 |
+
"eval_physics_loss": 0.9413084983825684,
|
| 1301 |
+
"eval_physics_mean_token_accuracy": 0.7854379539489746,
|
| 1302 |
+
"eval_physics_num_tokens": 29789253.0,
|
| 1303 |
+
"eval_physics_runtime": 84.3345,
|
| 1304 |
+
"eval_physics_samples_per_second": 5.929,
|
| 1305 |
+
"eval_physics_steps_per_second": 1.482,
|
| 1306 |
+
"step": 900
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"entropy": 0.8592871313914656,
|
| 1310 |
+
"epoch": 1.456,
|
| 1311 |
+
"grad_norm": 10.5,
|
| 1312 |
+
"learning_rate": 1.548888888888889e-05,
|
| 1313 |
+
"loss": 14.4092,
|
| 1314 |
+
"mean_token_accuracy": 0.7947866775095462,
|
| 1315 |
+
"num_tokens": 30128576.0,
|
| 1316 |
+
"step": 910
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"entropy": 0.9165356293320656,
|
| 1320 |
+
"epoch": 1.472,
|
| 1321 |
+
"grad_norm": 12.8125,
|
| 1322 |
+
"learning_rate": 1.5414814814814814e-05,
|
| 1323 |
+
"loss": 15.368,
|
| 1324 |
+
"mean_token_accuracy": 0.7829437594860792,
|
| 1325 |
+
"num_tokens": 30452719.0,
|
| 1326 |
+
"step": 920
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"entropy": 0.9012311644852161,
|
| 1330 |
+
"epoch": 1.488,
|
| 1331 |
+
"grad_norm": 10.9375,
|
| 1332 |
+
"learning_rate": 1.5340740740740744e-05,
|
| 1333 |
+
"loss": 14.9137,
|
| 1334 |
+
"mean_token_accuracy": 0.7879349775612354,
|
| 1335 |
+
"num_tokens": 30782343.0,
|
| 1336 |
+
"step": 930
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"entropy": 0.8793585889041424,
|
| 1340 |
+
"epoch": 1.504,
|
| 1341 |
+
"grad_norm": 11.75,
|
| 1342 |
+
"learning_rate": 1.5266666666666667e-05,
|
| 1343 |
+
"loss": 14.6612,
|
| 1344 |
+
"mean_token_accuracy": 0.7913552910089493,
|
| 1345 |
+
"num_tokens": 31113238.0,
|
| 1346 |
+
"step": 940
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"entropy": 0.896404268220067,
|
| 1350 |
+
"epoch": 1.52,
|
| 1351 |
+
"grad_norm": 12.25,
|
| 1352 |
+
"learning_rate": 1.5192592592592594e-05,
|
| 1353 |
+
"loss": 15.0177,
|
| 1354 |
+
"mean_token_accuracy": 0.7869206938892603,
|
| 1355 |
+
"num_tokens": 31444589.0,
|
| 1356 |
+
"step": 950
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"entropy": 0.9115363927558064,
|
| 1360 |
+
"epoch": 1.536,
|
| 1361 |
+
"grad_norm": 13.625,
|
| 1362 |
+
"learning_rate": 1.5118518518518519e-05,
|
| 1363 |
+
"loss": 15.1508,
|
| 1364 |
+
"mean_token_accuracy": 0.7850655887275935,
|
| 1365 |
+
"num_tokens": 31774137.0,
|
| 1366 |
+
"step": 960
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"entropy": 0.8715274870395661,
|
| 1370 |
+
"epoch": 1.552,
|
| 1371 |
+
"grad_norm": 12.5,
|
| 1372 |
+
"learning_rate": 1.5044444444444445e-05,
|
| 1373 |
+
"loss": 14.4724,
|
| 1374 |
+
"mean_token_accuracy": 0.7919801536947488,
|
| 1375 |
+
"num_tokens": 32101776.0,
|
| 1376 |
+
"step": 970
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"entropy": 0.8716113748028874,
|
| 1380 |
+
"epoch": 1.568,
|
| 1381 |
+
"grad_norm": 11.9375,
|
| 1382 |
+
"learning_rate": 1.497037037037037e-05,
|
| 1383 |
+
"loss": 14.5441,
|
| 1384 |
+
"mean_token_accuracy": 0.7928904049098492,
|
| 1385 |
+
"num_tokens": 32428021.0,
|
| 1386 |
+
"step": 980
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"entropy": 0.9078653456643224,
|
| 1390 |
+
"epoch": 1.584,
|
| 1391 |
+
"grad_norm": 12.375,
|
| 1392 |
+
"learning_rate": 1.4896296296296298e-05,
|
| 1393 |
+
"loss": 15.1365,
|
| 1394 |
+
"mean_token_accuracy": 0.7834811493754387,
|
| 1395 |
+
"num_tokens": 32752735.0,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"entropy": 0.8948801588267088,
|
| 1400 |
+
"epoch": 1.6,
|
| 1401 |
+
"grad_norm": 11.4375,
|
| 1402 |
+
"learning_rate": 1.4822222222222225e-05,
|
| 1403 |
+
"loss": 14.975,
|
| 1404 |
+
"mean_token_accuracy": 0.7872815534472466,
|
| 1405 |
+
"num_tokens": 33075822.0,
|
| 1406 |
+
"step": 1000
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 1.6,
|
| 1410 |
+
"eval_biology_entropy": 2.508749726295471,
|
| 1411 |
+
"eval_biology_loss": 2.8778915405273438,
|
| 1412 |
+
"eval_biology_mean_token_accuracy": 0.5310802006721497,
|
| 1413 |
+
"eval_biology_num_tokens": 33075822.0,
|
| 1414 |
+
"eval_biology_runtime": 57.246,
|
| 1415 |
+
"eval_biology_samples_per_second": 8.734,
|
| 1416 |
+
"eval_biology_steps_per_second": 2.184,
|
| 1417 |
+
"step": 1000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.6,
|
| 1421 |
+
"eval_chemistry_entropy": 1.2282180700302123,
|
| 1422 |
+
"eval_chemistry_loss": 1.2942099571228027,
|
| 1423 |
+
"eval_chemistry_mean_token_accuracy": 0.7318216099739074,
|
| 1424 |
+
"eval_chemistry_num_tokens": 33075822.0,
|
| 1425 |
+
"eval_chemistry_runtime": 71.0784,
|
| 1426 |
+
"eval_chemistry_samples_per_second": 7.034,
|
| 1427 |
+
"eval_chemistry_steps_per_second": 1.759,
|
| 1428 |
+
"step": 1000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 1.6,
|
| 1432 |
+
"eval_math_entropy": 0.9116221733093262,
|
| 1433 |
+
"eval_math_loss": 1.1501303911209106,
|
| 1434 |
+
"eval_math_mean_token_accuracy": 0.7557219748497009,
|
| 1435 |
+
"eval_math_num_tokens": 33075822.0,
|
| 1436 |
+
"eval_math_runtime": 74.8337,
|
| 1437 |
+
"eval_math_samples_per_second": 6.681,
|
| 1438 |
+
"eval_math_steps_per_second": 1.67,
|
| 1439 |
+
"step": 1000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.6,
|
| 1443 |
+
"eval_physics_entropy": 0.8932713055610657,
|
| 1444 |
+
"eval_physics_loss": 0.9365818500518799,
|
| 1445 |
+
"eval_physics_mean_token_accuracy": 0.785537106513977,
|
| 1446 |
+
"eval_physics_num_tokens": 33075822.0,
|
| 1447 |
+
"eval_physics_runtime": 84.6443,
|
| 1448 |
+
"eval_physics_samples_per_second": 5.907,
|
| 1449 |
+
"eval_physics_steps_per_second": 1.477,
|
| 1450 |
+
"step": 1000
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"entropy": 0.8604803189635277,
|
| 1454 |
+
"epoch": 1.616,
|
| 1455 |
+
"grad_norm": 12.5,
|
| 1456 |
+
"learning_rate": 1.474814814814815e-05,
|
| 1457 |
+
"loss": 14.3639,
|
| 1458 |
+
"mean_token_accuracy": 0.7951964545994997,
|
| 1459 |
+
"num_tokens": 33411850.0,
|
| 1460 |
+
"step": 1010
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"entropy": 0.8778028018772602,
|
| 1464 |
+
"epoch": 1.6320000000000001,
|
| 1465 |
+
"grad_norm": 12.5625,
|
| 1466 |
+
"learning_rate": 1.4674074074074076e-05,
|
| 1467 |
+
"loss": 14.6452,
|
| 1468 |
+
"mean_token_accuracy": 0.7921328764408827,
|
| 1469 |
+
"num_tokens": 33739456.0,
|
| 1470 |
+
"step": 1020
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"entropy": 0.853967827372253,
|
| 1474 |
+
"epoch": 1.6480000000000001,
|
| 1475 |
+
"grad_norm": 12.375,
|
| 1476 |
+
"learning_rate": 1.46e-05,
|
| 1477 |
+
"loss": 14.2504,
|
| 1478 |
+
"mean_token_accuracy": 0.7954076964408159,
|
| 1479 |
+
"num_tokens": 34072022.0,
|
| 1480 |
+
"step": 1030
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"entropy": 0.8785163436084986,
|
| 1484 |
+
"epoch": 1.6640000000000001,
|
| 1485 |
+
"grad_norm": 13.5625,
|
| 1486 |
+
"learning_rate": 1.4525925925925927e-05,
|
| 1487 |
+
"loss": 14.7032,
|
| 1488 |
+
"mean_token_accuracy": 0.7915361914783716,
|
| 1489 |
+
"num_tokens": 34398959.0,
|
| 1490 |
+
"step": 1040
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"entropy": 0.8694784231483936,
|
| 1494 |
+
"epoch": 1.6800000000000002,
|
| 1495 |
+
"grad_norm": 11.1875,
|
| 1496 |
+
"learning_rate": 1.4451851851851852e-05,
|
| 1497 |
+
"loss": 14.3839,
|
| 1498 |
+
"mean_token_accuracy": 0.7940326742827892,
|
| 1499 |
+
"num_tokens": 34732284.0,
|
| 1500 |
+
"step": 1050
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"entropy": 0.8717694684863091,
|
| 1504 |
+
"epoch": 1.696,
|
| 1505 |
+
"grad_norm": 13.1875,
|
| 1506 |
+
"learning_rate": 1.4377777777777779e-05,
|
| 1507 |
+
"loss": 14.6494,
|
| 1508 |
+
"mean_token_accuracy": 0.7915462471544743,
|
| 1509 |
+
"num_tokens": 35053712.0,
|
| 1510 |
+
"step": 1060
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"entropy": 0.8986424550414085,
|
| 1514 |
+
"epoch": 1.712,
|
| 1515 |
+
"grad_norm": 14.4375,
|
| 1516 |
+
"learning_rate": 1.4303703703703703e-05,
|
| 1517 |
+
"loss": 14.8976,
|
| 1518 |
+
"mean_token_accuracy": 0.7876383919268847,
|
| 1519 |
+
"num_tokens": 35393088.0,
|
| 1520 |
+
"step": 1070
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"entropy": 0.8717451239004731,
|
| 1524 |
+
"epoch": 1.728,
|
| 1525 |
+
"grad_norm": 12.6875,
|
| 1526 |
+
"learning_rate": 1.4229629629629632e-05,
|
| 1527 |
+
"loss": 14.5646,
|
| 1528 |
+
"mean_token_accuracy": 0.7918220609426498,
|
| 1529 |
+
"num_tokens": 35729007.0,
|
| 1530 |
+
"step": 1080
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"entropy": 0.8697011994197965,
|
| 1534 |
+
"epoch": 1.744,
|
| 1535 |
+
"grad_norm": 13.5,
|
| 1536 |
+
"learning_rate": 1.4155555555555556e-05,
|
| 1537 |
+
"loss": 14.5182,
|
| 1538 |
+
"mean_token_accuracy": 0.7922107793390751,
|
| 1539 |
+
"num_tokens": 36072489.0,
|
| 1540 |
+
"step": 1090
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"entropy": 0.8740223359316588,
|
| 1544 |
+
"epoch": 1.76,
|
| 1545 |
+
"grad_norm": 12.0,
|
| 1546 |
+
"learning_rate": 1.4081481481481483e-05,
|
| 1547 |
+
"loss": 14.4753,
|
| 1548 |
+
"mean_token_accuracy": 0.7926081418991089,
|
| 1549 |
+
"num_tokens": 36398285.0,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.76,
|
| 1554 |
+
"eval_biology_entropy": 2.487686212539673,
|
| 1555 |
+
"eval_biology_loss": 2.874035120010376,
|
| 1556 |
+
"eval_biology_mean_token_accuracy": 0.5334258432388306,
|
| 1557 |
+
"eval_biology_num_tokens": 36398285.0,
|
| 1558 |
+
"eval_biology_runtime": 57.5039,
|
| 1559 |
+
"eval_biology_samples_per_second": 8.695,
|
| 1560 |
+
"eval_biology_steps_per_second": 2.174,
|
| 1561 |
+
"step": 1100
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.76,
|
| 1565 |
+
"eval_chemistry_entropy": 1.2144387559890748,
|
| 1566 |
+
"eval_chemistry_loss": 1.29018235206604,
|
| 1567 |
+
"eval_chemistry_mean_token_accuracy": 0.7329304647445679,
|
| 1568 |
+
"eval_chemistry_num_tokens": 36398285.0,
|
| 1569 |
+
"eval_chemistry_runtime": 71.5726,
|
| 1570 |
+
"eval_chemistry_samples_per_second": 6.986,
|
| 1571 |
+
"eval_chemistry_steps_per_second": 1.746,
|
| 1572 |
+
"step": 1100
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 1.76,
|
| 1576 |
+
"eval_math_entropy": 0.9051191382408142,
|
| 1577 |
+
"eval_math_loss": 1.1472207307815552,
|
| 1578 |
+
"eval_math_mean_token_accuracy": 0.7566594734191895,
|
| 1579 |
+
"eval_math_num_tokens": 36398285.0,
|
| 1580 |
+
"eval_math_runtime": 74.0714,
|
| 1581 |
+
"eval_math_samples_per_second": 6.75,
|
| 1582 |
+
"eval_math_steps_per_second": 1.688,
|
| 1583 |
+
"step": 1100
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 1.76,
|
| 1587 |
+
"eval_physics_entropy": 0.8845951933860778,
|
| 1588 |
+
"eval_physics_loss": 0.9326092600822449,
|
| 1589 |
+
"eval_physics_mean_token_accuracy": 0.7863762426376343,
|
| 1590 |
+
"eval_physics_num_tokens": 36398285.0,
|
| 1591 |
+
"eval_physics_runtime": 84.7278,
|
| 1592 |
+
"eval_physics_samples_per_second": 5.901,
|
| 1593 |
+
"eval_physics_steps_per_second": 1.475,
|
| 1594 |
+
"step": 1100
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"entropy": 0.9020034002140165,
|
| 1598 |
+
"epoch": 1.776,
|
| 1599 |
+
"grad_norm": 12.625,
|
| 1600 |
+
"learning_rate": 1.400740740740741e-05,
|
| 1601 |
+
"loss": 15.0724,
|
| 1602 |
+
"mean_token_accuracy": 0.7856271650642157,
|
| 1603 |
+
"num_tokens": 36730479.0,
|
| 1604 |
+
"step": 1110
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"entropy": 0.862918908149004,
|
| 1608 |
+
"epoch": 1.792,
|
| 1609 |
+
"grad_norm": 14.6875,
|
| 1610 |
+
"learning_rate": 1.3933333333333334e-05,
|
| 1611 |
+
"loss": 14.5005,
|
| 1612 |
+
"mean_token_accuracy": 0.7926642511039972,
|
| 1613 |
+
"num_tokens": 37057425.0,
|
| 1614 |
+
"step": 1120
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"entropy": 0.8617109794169664,
|
| 1618 |
+
"epoch": 1.808,
|
| 1619 |
+
"grad_norm": 11.8125,
|
| 1620 |
+
"learning_rate": 1.385925925925926e-05,
|
| 1621 |
+
"loss": 14.2625,
|
| 1622 |
+
"mean_token_accuracy": 0.7952517606317997,
|
| 1623 |
+
"num_tokens": 37395449.0,
|
| 1624 |
+
"step": 1130
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"entropy": 0.8684184370562434,
|
| 1628 |
+
"epoch": 1.8239999999999998,
|
| 1629 |
+
"grad_norm": 12.5,
|
| 1630 |
+
"learning_rate": 1.3785185185185186e-05,
|
| 1631 |
+
"loss": 14.5241,
|
| 1632 |
+
"mean_token_accuracy": 0.7931536667048931,
|
| 1633 |
+
"num_tokens": 37729535.0,
|
| 1634 |
+
"step": 1140
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"entropy": 0.8654272655025125,
|
| 1638 |
+
"epoch": 1.8399999999999999,
|
| 1639 |
+
"grad_norm": 12.4375,
|
| 1640 |
+
"learning_rate": 1.3711111111111112e-05,
|
| 1641 |
+
"loss": 14.4054,
|
| 1642 |
+
"mean_token_accuracy": 0.7938401244580746,
|
| 1643 |
+
"num_tokens": 38057839.0,
|
| 1644 |
+
"step": 1150
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"entropy": 0.86546247061342,
|
| 1648 |
+
"epoch": 1.8559999999999999,
|
| 1649 |
+
"grad_norm": 13.4375,
|
| 1650 |
+
"learning_rate": 1.3637037037037037e-05,
|
| 1651 |
+
"loss": 14.4546,
|
| 1652 |
+
"mean_token_accuracy": 0.7931617993861437,
|
| 1653 |
+
"num_tokens": 38398033.0,
|
| 1654 |
+
"step": 1160
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"entropy": 0.8690170094370842,
|
| 1658 |
+
"epoch": 1.8719999999999999,
|
| 1659 |
+
"grad_norm": 11.8125,
|
| 1660 |
+
"learning_rate": 1.3562962962962965e-05,
|
| 1661 |
+
"loss": 14.4475,
|
| 1662 |
+
"mean_token_accuracy": 0.7915640283375979,
|
| 1663 |
+
"num_tokens": 38729203.0,
|
| 1664 |
+
"step": 1170
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"entropy": 0.8866201037541032,
|
| 1668 |
+
"epoch": 1.888,
|
| 1669 |
+
"grad_norm": 13.0625,
|
| 1670 |
+
"learning_rate": 1.3488888888888888e-05,
|
| 1671 |
+
"loss": 14.8192,
|
| 1672 |
+
"mean_token_accuracy": 0.7894200544804335,
|
| 1673 |
+
"num_tokens": 39058791.0,
|
| 1674 |
+
"step": 1180
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"entropy": 0.8984544143080712,
|
| 1678 |
+
"epoch": 1.904,
|
| 1679 |
+
"grad_norm": 12.25,
|
| 1680 |
+
"learning_rate": 1.3414814814814817e-05,
|
| 1681 |
+
"loss": 15.0873,
|
| 1682 |
+
"mean_token_accuracy": 0.7861377280205488,
|
| 1683 |
+
"num_tokens": 39386170.0,
|
| 1684 |
+
"step": 1190
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"entropy": 0.9040035644546152,
|
| 1688 |
+
"epoch": 1.92,
|
| 1689 |
+
"grad_norm": 12.0625,
|
| 1690 |
+
"learning_rate": 1.3340740740740741e-05,
|
| 1691 |
+
"loss": 14.9941,
|
| 1692 |
+
"mean_token_accuracy": 0.7868796121329069,
|
| 1693 |
+
"num_tokens": 39710381.0,
|
| 1694 |
+
"step": 1200
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 1.92,
|
| 1698 |
+
"eval_biology_entropy": 2.485040246009827,
|
| 1699 |
+
"eval_biology_loss": 2.8711678981781006,
|
| 1700 |
+
"eval_biology_mean_token_accuracy": 0.5342430763244629,
|
| 1701 |
+
"eval_biology_num_tokens": 39710381.0,
|
| 1702 |
+
"eval_biology_runtime": 57.5639,
|
| 1703 |
+
"eval_biology_samples_per_second": 8.686,
|
| 1704 |
+
"eval_biology_steps_per_second": 2.171,
|
| 1705 |
+
"step": 1200
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 1.92,
|
| 1709 |
+
"eval_chemistry_entropy": 1.208739689350128,
|
| 1710 |
+
"eval_chemistry_loss": 1.285979151725769,
|
| 1711 |
+
"eval_chemistry_mean_token_accuracy": 0.7334310331344605,
|
| 1712 |
+
"eval_chemistry_num_tokens": 39710381.0,
|
| 1713 |
+
"eval_chemistry_runtime": 71.5821,
|
| 1714 |
+
"eval_chemistry_samples_per_second": 6.985,
|
| 1715 |
+
"eval_chemistry_steps_per_second": 1.746,
|
| 1716 |
+
"step": 1200
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 1.92,
|
| 1720 |
+
"eval_math_entropy": 0.9018509998321533,
|
| 1721 |
+
"eval_math_loss": 1.1460354328155518,
|
| 1722 |
+
"eval_math_mean_token_accuracy": 0.7567954559326172,
|
| 1723 |
+
"eval_math_num_tokens": 39710381.0,
|
| 1724 |
+
"eval_math_runtime": 75.1131,
|
| 1725 |
+
"eval_math_samples_per_second": 6.657,
|
| 1726 |
+
"eval_math_steps_per_second": 1.664,
|
| 1727 |
+
"step": 1200
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 1.92,
|
| 1731 |
+
"eval_physics_entropy": 0.8798905625343323,
|
| 1732 |
+
"eval_physics_loss": 0.929322361946106,
|
| 1733 |
+
"eval_physics_mean_token_accuracy": 0.787430817604065,
|
| 1734 |
+
"eval_physics_num_tokens": 39710381.0,
|
| 1735 |
+
"eval_physics_runtime": 84.784,
|
| 1736 |
+
"eval_physics_samples_per_second": 5.897,
|
| 1737 |
+
"eval_physics_steps_per_second": 1.474,
|
| 1738 |
+
"step": 1200
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"entropy": 0.8708227036520839,
|
| 1742 |
+
"epoch": 1.936,
|
| 1743 |
+
"grad_norm": 13.8125,
|
| 1744 |
+
"learning_rate": 1.3266666666666668e-05,
|
| 1745 |
+
"loss": 14.626,
|
| 1746 |
+
"mean_token_accuracy": 0.7907645083963871,
|
| 1747 |
+
"num_tokens": 40043354.0,
|
| 1748 |
+
"step": 1210
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"entropy": 0.8374214060604572,
|
| 1752 |
+
"epoch": 1.952,
|
| 1753 |
+
"grad_norm": 11.5,
|
| 1754 |
+
"learning_rate": 1.3192592592592594e-05,
|
| 1755 |
+
"loss": 13.9343,
|
| 1756 |
+
"mean_token_accuracy": 0.7994464814662934,
|
| 1757 |
+
"num_tokens": 40375790.0,
|
| 1758 |
+
"step": 1220
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"entropy": 0.8776708476245403,
|
| 1762 |
+
"epoch": 1.968,
|
| 1763 |
+
"grad_norm": 12.25,
|
| 1764 |
+
"learning_rate": 1.311851851851852e-05,
|
| 1765 |
+
"loss": 14.5653,
|
| 1766 |
+
"mean_token_accuracy": 0.7916903175413609,
|
| 1767 |
+
"num_tokens": 40705451.0,
|
| 1768 |
+
"step": 1230
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"entropy": 0.8854383405297994,
|
| 1772 |
+
"epoch": 1.984,
|
| 1773 |
+
"grad_norm": 13.0,
|
| 1774 |
+
"learning_rate": 1.3044444444444446e-05,
|
| 1775 |
+
"loss": 14.7486,
|
| 1776 |
+
"mean_token_accuracy": 0.7898764718323946,
|
| 1777 |
+
"num_tokens": 41040313.0,
|
| 1778 |
+
"step": 1240
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"entropy": 0.879773647710681,
|
| 1782 |
+
"epoch": 2.0,
|
| 1783 |
+
"grad_norm": 12.6875,
|
| 1784 |
+
"learning_rate": 1.297037037037037e-05,
|
| 1785 |
+
"loss": 14.6916,
|
| 1786 |
+
"mean_token_accuracy": 0.7911134500056505,
|
| 1787 |
+
"num_tokens": 41373608.0,
|
| 1788 |
+
"step": 1250
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"entropy": 0.8601289037615061,
|
| 1792 |
+
"epoch": 2.016,
|
| 1793 |
+
"grad_norm": 10.5625,
|
| 1794 |
+
"learning_rate": 1.2896296296296299e-05,
|
| 1795 |
+
"loss": 14.1908,
|
| 1796 |
+
"mean_token_accuracy": 0.7955393616110087,
|
| 1797 |
+
"num_tokens": 41708226.0,
|
| 1798 |
+
"step": 1260
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"entropy": 0.8542119370773434,
|
| 1802 |
+
"epoch": 2.032,
|
| 1803 |
+
"grad_norm": 11.4375,
|
| 1804 |
+
"learning_rate": 1.2822222222222222e-05,
|
| 1805 |
+
"loss": 14.2867,
|
| 1806 |
+
"mean_token_accuracy": 0.7959718242287636,
|
| 1807 |
+
"num_tokens": 42045433.0,
|
| 1808 |
+
"step": 1270
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"entropy": 0.8921715309843421,
|
| 1812 |
+
"epoch": 2.048,
|
| 1813 |
+
"grad_norm": 13.1875,
|
| 1814 |
+
"learning_rate": 1.274814814814815e-05,
|
| 1815 |
+
"loss": 14.8772,
|
| 1816 |
+
"mean_token_accuracy": 0.7874497707933188,
|
| 1817 |
+
"num_tokens": 42372694.0,
|
| 1818 |
+
"step": 1280
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"entropy": 0.8619997894391418,
|
| 1822 |
+
"epoch": 2.064,
|
| 1823 |
+
"grad_norm": 12.375,
|
| 1824 |
+
"learning_rate": 1.2674074074074075e-05,
|
| 1825 |
+
"loss": 14.2556,
|
| 1826 |
+
"mean_token_accuracy": 0.7947564154863358,
|
| 1827 |
+
"num_tokens": 42710240.0,
|
| 1828 |
+
"step": 1290
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"entropy": 0.8751549379900098,
|
| 1832 |
+
"epoch": 2.08,
|
| 1833 |
+
"grad_norm": 11.5625,
|
| 1834 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 1835 |
+
"loss": 14.6573,
|
| 1836 |
+
"mean_token_accuracy": 0.7916811484843492,
|
| 1837 |
+
"num_tokens": 43037844.0,
|
| 1838 |
+
"step": 1300
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 2.08,
|
| 1842 |
+
"eval_biology_entropy": 2.5093424930572508,
|
| 1843 |
+
"eval_biology_loss": 2.864481210708618,
|
| 1844 |
+
"eval_biology_mean_token_accuracy": 0.533148678779602,
|
| 1845 |
+
"eval_biology_num_tokens": 43037844.0,
|
| 1846 |
+
"eval_biology_runtime": 57.534,
|
| 1847 |
+
"eval_biology_samples_per_second": 8.691,
|
| 1848 |
+
"eval_biology_steps_per_second": 2.173,
|
| 1849 |
+
"step": 1300
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 2.08,
|
| 1853 |
+
"eval_chemistry_entropy": 1.2106250019073486,
|
| 1854 |
+
"eval_chemistry_loss": 1.2846547365188599,
|
| 1855 |
+
"eval_chemistry_mean_token_accuracy": 0.7332292509078979,
|
| 1856 |
+
"eval_chemistry_num_tokens": 43037844.0,
|
| 1857 |
+
"eval_chemistry_runtime": 71.3949,
|
| 1858 |
+
"eval_chemistry_samples_per_second": 7.003,
|
| 1859 |
+
"eval_chemistry_steps_per_second": 1.751,
|
| 1860 |
+
"step": 1300
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 2.08,
|
| 1864 |
+
"eval_math_entropy": 0.9025118923187256,
|
| 1865 |
+
"eval_math_loss": 1.1455634832382202,
|
| 1866 |
+
"eval_math_mean_token_accuracy": 0.7569516344070435,
|
| 1867 |
+
"eval_math_num_tokens": 43037844.0,
|
| 1868 |
+
"eval_math_runtime": 75.2696,
|
| 1869 |
+
"eval_math_samples_per_second": 6.643,
|
| 1870 |
+
"eval_math_steps_per_second": 1.661,
|
| 1871 |
+
"step": 1300
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.08,
|
| 1875 |
+
"eval_physics_entropy": 0.8776944227218628,
|
| 1876 |
+
"eval_physics_loss": 0.9265705943107605,
|
| 1877 |
+
"eval_physics_mean_token_accuracy": 0.7875892186164856,
|
| 1878 |
+
"eval_physics_num_tokens": 43037844.0,
|
| 1879 |
+
"eval_physics_runtime": 85.0265,
|
| 1880 |
+
"eval_physics_samples_per_second": 5.881,
|
| 1881 |
+
"eval_physics_steps_per_second": 1.47,
|
| 1882 |
+
"step": 1300
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"entropy": 0.8755308095365762,
|
| 1886 |
+
"epoch": 2.096,
|
| 1887 |
+
"grad_norm": 12.9375,
|
| 1888 |
+
"learning_rate": 1.2525925925925928e-05,
|
| 1889 |
+
"loss": 14.5668,
|
| 1890 |
+
"mean_token_accuracy": 0.7918181721121073,
|
| 1891 |
+
"num_tokens": 43368675.0,
|
| 1892 |
+
"step": 1310
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"entropy": 0.8589686010032892,
|
| 1896 |
+
"epoch": 2.112,
|
| 1897 |
+
"grad_norm": 12.25,
|
| 1898 |
+
"learning_rate": 1.2451851851851853e-05,
|
| 1899 |
+
"loss": 14.3302,
|
| 1900 |
+
"mean_token_accuracy": 0.7948925741016865,
|
| 1901 |
+
"num_tokens": 43699544.0,
|
| 1902 |
+
"step": 1320
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"entropy": 0.8743958847597242,
|
| 1906 |
+
"epoch": 2.128,
|
| 1907 |
+
"grad_norm": 12.3125,
|
| 1908 |
+
"learning_rate": 1.237777777777778e-05,
|
| 1909 |
+
"loss": 14.5493,
|
| 1910 |
+
"mean_token_accuracy": 0.7907001547515392,
|
| 1911 |
+
"num_tokens": 44012632.0,
|
| 1912 |
+
"step": 1330
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"entropy": 0.8594208642840385,
|
| 1916 |
+
"epoch": 2.144,
|
| 1917 |
+
"grad_norm": 11.0625,
|
| 1918 |
+
"learning_rate": 1.2303703703703704e-05,
|
| 1919 |
+
"loss": 14.3324,
|
| 1920 |
+
"mean_token_accuracy": 0.7929412983357906,
|
| 1921 |
+
"num_tokens": 44349732.0,
|
| 1922 |
+
"step": 1340
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"entropy": 0.8937922870740295,
|
| 1926 |
+
"epoch": 2.16,
|
| 1927 |
+
"grad_norm": 12.625,
|
| 1928 |
+
"learning_rate": 1.222962962962963e-05,
|
| 1929 |
+
"loss": 14.8834,
|
| 1930 |
+
"mean_token_accuracy": 0.7873219456523657,
|
| 1931 |
+
"num_tokens": 44680410.0,
|
| 1932 |
+
"step": 1350
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"entropy": 0.8523136384785175,
|
| 1936 |
+
"epoch": 2.176,
|
| 1937 |
+
"grad_norm": 11.25,
|
| 1938 |
+
"learning_rate": 1.2155555555555555e-05,
|
| 1939 |
+
"loss": 14.2544,
|
| 1940 |
+
"mean_token_accuracy": 0.7959212839603425,
|
| 1941 |
+
"num_tokens": 45017922.0,
|
| 1942 |
+
"step": 1360
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"entropy": 0.8626351818442345,
|
| 1946 |
+
"epoch": 2.192,
|
| 1947 |
+
"grad_norm": 12.1875,
|
| 1948 |
+
"learning_rate": 1.2081481481481484e-05,
|
| 1949 |
+
"loss": 14.276,
|
| 1950 |
+
"mean_token_accuracy": 0.7951519623398781,
|
| 1951 |
+
"num_tokens": 45348179.0,
|
| 1952 |
+
"step": 1370
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"entropy": 0.8525467690080404,
|
| 1956 |
+
"epoch": 2.208,
|
| 1957 |
+
"grad_norm": 12.8125,
|
| 1958 |
+
"learning_rate": 1.2007407407407408e-05,
|
| 1959 |
+
"loss": 14.2013,
|
| 1960 |
+
"mean_token_accuracy": 0.7966051142662763,
|
| 1961 |
+
"num_tokens": 45678397.0,
|
| 1962 |
+
"step": 1380
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"entropy": 0.8527483340352774,
|
| 1966 |
+
"epoch": 2.224,
|
| 1967 |
+
"grad_norm": 11.5625,
|
| 1968 |
+
"learning_rate": 1.1933333333333335e-05,
|
| 1969 |
+
"loss": 14.2627,
|
| 1970 |
+
"mean_token_accuracy": 0.7949573740363121,
|
| 1971 |
+
"num_tokens": 46010052.0,
|
| 1972 |
+
"step": 1390
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"entropy": 0.855360858142376,
|
| 1976 |
+
"epoch": 2.24,
|
| 1977 |
+
"grad_norm": 11.4375,
|
| 1978 |
+
"learning_rate": 1.185925925925926e-05,
|
| 1979 |
+
"loss": 14.2457,
|
| 1980 |
+
"mean_token_accuracy": 0.7951466862112284,
|
| 1981 |
+
"num_tokens": 46347116.0,
|
| 1982 |
+
"step": 1400
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 2.24,
|
| 1986 |
+
"eval_biology_entropy": 2.5089185886383056,
|
| 1987 |
+
"eval_biology_loss": 2.863374948501587,
|
| 1988 |
+
"eval_biology_mean_token_accuracy": 0.5349591159820557,
|
| 1989 |
+
"eval_biology_num_tokens": 46347116.0,
|
| 1990 |
+
"eval_biology_runtime": 57.3344,
|
| 1991 |
+
"eval_biology_samples_per_second": 8.721,
|
| 1992 |
+
"eval_biology_steps_per_second": 2.18,
|
| 1993 |
+
"step": 1400
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 2.24,
|
| 1997 |
+
"eval_chemistry_entropy": 1.2185512976646424,
|
| 1998 |
+
"eval_chemistry_loss": 1.2818259000778198,
|
| 1999 |
+
"eval_chemistry_mean_token_accuracy": 0.7342271280288696,
|
| 2000 |
+
"eval_chemistry_num_tokens": 46347116.0,
|
| 2001 |
+
"eval_chemistry_runtime": 71.4639,
|
| 2002 |
+
"eval_chemistry_samples_per_second": 6.997,
|
| 2003 |
+
"eval_chemistry_steps_per_second": 1.749,
|
| 2004 |
+
"step": 1400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 2.24,
|
| 2008 |
+
"eval_math_entropy": 0.9067861828804016,
|
| 2009 |
+
"eval_math_loss": 1.1431993246078491,
|
| 2010 |
+
"eval_math_mean_token_accuracy": 0.7572210087776184,
|
| 2011 |
+
"eval_math_num_tokens": 46347116.0,
|
| 2012 |
+
"eval_math_runtime": 75.3577,
|
| 2013 |
+
"eval_math_samples_per_second": 6.635,
|
| 2014 |
+
"eval_math_steps_per_second": 1.659,
|
| 2015 |
+
"step": 1400
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 2.24,
|
| 2019 |
+
"eval_physics_entropy": 0.8840774774551392,
|
| 2020 |
+
"eval_physics_loss": 0.9245942234992981,
|
| 2021 |
+
"eval_physics_mean_token_accuracy": 0.7878187990188599,
|
| 2022 |
+
"eval_physics_num_tokens": 46347116.0,
|
| 2023 |
+
"eval_physics_runtime": 84.9011,
|
| 2024 |
+
"eval_physics_samples_per_second": 5.889,
|
| 2025 |
+
"eval_physics_steps_per_second": 1.472,
|
| 2026 |
+
"step": 1400
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"entropy": 0.9004593985155225,
|
| 2030 |
+
"epoch": 2.2560000000000002,
|
| 2031 |
+
"grad_norm": 13.4375,
|
| 2032 |
+
"learning_rate": 1.1785185185185186e-05,
|
| 2033 |
+
"loss": 14.9385,
|
| 2034 |
+
"mean_token_accuracy": 0.7881197012960911,
|
| 2035 |
+
"num_tokens": 46673074.0,
|
| 2036 |
+
"step": 1410
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"entropy": 0.8454325065016747,
|
| 2040 |
+
"epoch": 2.2720000000000002,
|
| 2041 |
+
"grad_norm": 12.125,
|
| 2042 |
+
"learning_rate": 1.1711111111111113e-05,
|
| 2043 |
+
"loss": 14.0933,
|
| 2044 |
+
"mean_token_accuracy": 0.7979227758944034,
|
| 2045 |
+
"num_tokens": 47009386.0,
|
| 2046 |
+
"step": 1420
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"entropy": 0.8827449329197407,
|
| 2050 |
+
"epoch": 2.288,
|
| 2051 |
+
"grad_norm": 12.25,
|
| 2052 |
+
"learning_rate": 1.1637037037037037e-05,
|
| 2053 |
+
"loss": 14.712,
|
| 2054 |
+
"mean_token_accuracy": 0.7902342595160008,
|
| 2055 |
+
"num_tokens": 47344231.0,
|
| 2056 |
+
"step": 1430
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"entropy": 0.8806605780497193,
|
| 2060 |
+
"epoch": 2.304,
|
| 2061 |
+
"grad_norm": 13.4375,
|
| 2062 |
+
"learning_rate": 1.1562962962962964e-05,
|
| 2063 |
+
"loss": 14.5674,
|
| 2064 |
+
"mean_token_accuracy": 0.7924861270934344,
|
| 2065 |
+
"num_tokens": 47658392.0,
|
| 2066 |
+
"step": 1440
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"entropy": 0.859540069103241,
|
| 2070 |
+
"epoch": 2.32,
|
| 2071 |
+
"grad_norm": 12.1875,
|
| 2072 |
+
"learning_rate": 1.1488888888888889e-05,
|
| 2073 |
+
"loss": 14.3445,
|
| 2074 |
+
"mean_token_accuracy": 0.7936220798641443,
|
| 2075 |
+
"num_tokens": 47995015.0,
|
| 2076 |
+
"step": 1450
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"entropy": 0.8417845372110605,
|
| 2080 |
+
"epoch": 2.336,
|
| 2081 |
+
"grad_norm": 11.8125,
|
| 2082 |
+
"learning_rate": 1.1414814814814817e-05,
|
| 2083 |
+
"loss": 14.0115,
|
| 2084 |
+
"mean_token_accuracy": 0.7978896964341402,
|
| 2085 |
+
"num_tokens": 48336708.0,
|
| 2086 |
+
"step": 1460
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"entropy": 0.8409199349582195,
|
| 2090 |
+
"epoch": 2.352,
|
| 2091 |
+
"grad_norm": 11.5,
|
| 2092 |
+
"learning_rate": 1.1340740740740742e-05,
|
| 2093 |
+
"loss": 13.9932,
|
| 2094 |
+
"mean_token_accuracy": 0.7971417736262083,
|
| 2095 |
+
"num_tokens": 48665311.0,
|
| 2096 |
+
"step": 1470
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"entropy": 0.8575670935213566,
|
| 2100 |
+
"epoch": 2.368,
|
| 2101 |
+
"grad_norm": 11.75,
|
| 2102 |
+
"learning_rate": 1.1266666666666668e-05,
|
| 2103 |
+
"loss": 14.3902,
|
| 2104 |
+
"mean_token_accuracy": 0.7932044431567192,
|
| 2105 |
+
"num_tokens": 49005334.0,
|
| 2106 |
+
"step": 1480
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"entropy": 0.8846392493695021,
|
| 2110 |
+
"epoch": 2.384,
|
| 2111 |
+
"grad_norm": 13.5,
|
| 2112 |
+
"learning_rate": 1.1192592592592593e-05,
|
| 2113 |
+
"loss": 14.5912,
|
| 2114 |
+
"mean_token_accuracy": 0.7901103302836419,
|
| 2115 |
+
"num_tokens": 49331936.0,
|
| 2116 |
+
"step": 1490
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"entropy": 0.8669425016269088,
|
| 2120 |
+
"epoch": 2.4,
|
| 2121 |
+
"grad_norm": 12.4375,
|
| 2122 |
+
"learning_rate": 1.111851851851852e-05,
|
| 2123 |
+
"loss": 14.4765,
|
| 2124 |
+
"mean_token_accuracy": 0.7921019058674574,
|
| 2125 |
+
"num_tokens": 49671048.0,
|
| 2126 |
+
"step": 1500
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 2.4,
|
| 2130 |
+
"eval_biology_entropy": 2.456291440010071,
|
| 2131 |
+
"eval_biology_loss": 2.8605849742889404,
|
| 2132 |
+
"eval_biology_mean_token_accuracy": 0.5338850898742675,
|
| 2133 |
+
"eval_biology_num_tokens": 49671048.0,
|
| 2134 |
+
"eval_biology_runtime": 57.1011,
|
| 2135 |
+
"eval_biology_samples_per_second": 8.756,
|
| 2136 |
+
"eval_biology_steps_per_second": 2.189,
|
| 2137 |
+
"step": 1500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.4,
|
| 2141 |
+
"eval_chemistry_entropy": 1.1939663624763488,
|
| 2142 |
+
"eval_chemistry_loss": 1.280471682548523,
|
| 2143 |
+
"eval_chemistry_mean_token_accuracy": 0.7337013077735901,
|
| 2144 |
+
"eval_chemistry_num_tokens": 49671048.0,
|
| 2145 |
+
"eval_chemistry_runtime": 71.0078,
|
| 2146 |
+
"eval_chemistry_samples_per_second": 7.041,
|
| 2147 |
+
"eval_chemistry_steps_per_second": 1.76,
|
| 2148 |
+
"step": 1500
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 2.4,
|
| 2152 |
+
"eval_math_entropy": 0.8943547616004944,
|
| 2153 |
+
"eval_math_loss": 1.1426331996917725,
|
| 2154 |
+
"eval_math_mean_token_accuracy": 0.7575869603157044,
|
| 2155 |
+
"eval_math_num_tokens": 49671048.0,
|
| 2156 |
+
"eval_math_runtime": 74.2145,
|
| 2157 |
+
"eval_math_samples_per_second": 6.737,
|
| 2158 |
+
"eval_math_steps_per_second": 1.684,
|
| 2159 |
+
"step": 1500
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.4,
|
| 2163 |
+
"eval_physics_entropy": 0.8692562022209167,
|
| 2164 |
+
"eval_physics_loss": 0.9228239059448242,
|
| 2165 |
+
"eval_physics_mean_token_accuracy": 0.7882768630981445,
|
| 2166 |
+
"eval_physics_num_tokens": 49671048.0,
|
| 2167 |
+
"eval_physics_runtime": 85.556,
|
| 2168 |
+
"eval_physics_samples_per_second": 5.844,
|
| 2169 |
+
"eval_physics_steps_per_second": 1.461,
|
| 2170 |
+
"step": 1500
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"entropy": 0.8500566266477108,
|
| 2174 |
+
"epoch": 2.416,
|
| 2175 |
+
"grad_norm": 11.4375,
|
| 2176 |
+
"learning_rate": 1.1044444444444444e-05,
|
| 2177 |
+
"loss": 14.158,
|
| 2178 |
+
"mean_token_accuracy": 0.7961105518043041,
|
| 2179 |
+
"num_tokens": 50018182.0,
|
| 2180 |
+
"step": 1510
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"entropy": 0.8643727529793978,
|
| 2184 |
+
"epoch": 2.432,
|
| 2185 |
+
"grad_norm": 12.75,
|
| 2186 |
+
"learning_rate": 1.0970370370370371e-05,
|
| 2187 |
+
"loss": 14.4046,
|
| 2188 |
+
"mean_token_accuracy": 0.7947129063308239,
|
| 2189 |
+
"num_tokens": 50341187.0,
|
| 2190 |
+
"step": 1520
|
| 2191 |
+
},
|
| 2192 |
+
{
|
| 2193 |
+
"entropy": 0.8951457351446152,
|
| 2194 |
+
"epoch": 2.448,
|
| 2195 |
+
"grad_norm": 11.25,
|
| 2196 |
+
"learning_rate": 1.0896296296296298e-05,
|
| 2197 |
+
"loss": 14.8649,
|
| 2198 |
+
"mean_token_accuracy": 0.7867488101124763,
|
| 2199 |
+
"num_tokens": 50667019.0,
|
| 2200 |
+
"step": 1530
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"entropy": 0.8816273506730795,
|
| 2204 |
+
"epoch": 2.464,
|
| 2205 |
+
"grad_norm": 12.8125,
|
| 2206 |
+
"learning_rate": 1.0822222222222222e-05,
|
| 2207 |
+
"loss": 14.6,
|
| 2208 |
+
"mean_token_accuracy": 0.7916153628379107,
|
| 2209 |
+
"num_tokens": 50991113.0,
|
| 2210 |
+
"step": 1540
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"entropy": 0.8664202645421029,
|
| 2214 |
+
"epoch": 2.48,
|
| 2215 |
+
"grad_norm": 12.3125,
|
| 2216 |
+
"learning_rate": 1.074814814814815e-05,
|
| 2217 |
+
"loss": 14.5346,
|
| 2218 |
+
"mean_token_accuracy": 0.7921106025576592,
|
| 2219 |
+
"num_tokens": 51319624.0,
|
| 2220 |
+
"step": 1550
|
| 2221 |
+
},
|
| 2222 |
+
{
|
| 2223 |
+
"entropy": 0.8818996708840132,
|
| 2224 |
+
"epoch": 2.496,
|
| 2225 |
+
"grad_norm": 11.625,
|
| 2226 |
+
"learning_rate": 1.0674074074074074e-05,
|
| 2227 |
+
"loss": 14.6185,
|
| 2228 |
+
"mean_token_accuracy": 0.7901674829423427,
|
| 2229 |
+
"num_tokens": 51654633.0,
|
| 2230 |
+
"step": 1560
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"entropy": 0.8544634876772761,
|
| 2234 |
+
"epoch": 2.512,
|
| 2235 |
+
"grad_norm": 13.25,
|
| 2236 |
+
"learning_rate": 1.0600000000000002e-05,
|
| 2237 |
+
"loss": 14.2895,
|
| 2238 |
+
"mean_token_accuracy": 0.7949651554226875,
|
| 2239 |
+
"num_tokens": 51980051.0,
|
| 2240 |
+
"step": 1570
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"entropy": 0.8599010031670332,
|
| 2244 |
+
"epoch": 2.528,
|
| 2245 |
+
"grad_norm": 11.5625,
|
| 2246 |
+
"learning_rate": 1.0525925925925927e-05,
|
| 2247 |
+
"loss": 14.2789,
|
| 2248 |
+
"mean_token_accuracy": 0.7945577628910542,
|
| 2249 |
+
"num_tokens": 52316109.0,
|
| 2250 |
+
"step": 1580
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"entropy": 0.8746189616620541,
|
| 2254 |
+
"epoch": 2.544,
|
| 2255 |
+
"grad_norm": 13.6875,
|
| 2256 |
+
"learning_rate": 1.0451851851851853e-05,
|
| 2257 |
+
"loss": 14.6318,
|
| 2258 |
+
"mean_token_accuracy": 0.7903719995170831,
|
| 2259 |
+
"num_tokens": 52634528.0,
|
| 2260 |
+
"step": 1590
|
| 2261 |
+
},
|
| 2262 |
+
{
|
| 2263 |
+
"entropy": 0.8605435255914926,
|
| 2264 |
+
"epoch": 2.56,
|
| 2265 |
+
"grad_norm": 12.0,
|
| 2266 |
+
"learning_rate": 1.0377777777777778e-05,
|
| 2267 |
+
"loss": 14.3017,
|
| 2268 |
+
"mean_token_accuracy": 0.7939822390675545,
|
| 2269 |
+
"num_tokens": 52964035.0,
|
| 2270 |
+
"step": 1600
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 2.56,
|
| 2274 |
+
"eval_biology_entropy": 2.439929902076721,
|
| 2275 |
+
"eval_biology_loss": 2.855186939239502,
|
| 2276 |
+
"eval_biology_mean_token_accuracy": 0.5345207722187042,
|
| 2277 |
+
"eval_biology_num_tokens": 52964035.0,
|
| 2278 |
+
"eval_biology_runtime": 57.4094,
|
| 2279 |
+
"eval_biology_samples_per_second": 8.709,
|
| 2280 |
+
"eval_biology_steps_per_second": 2.177,
|
| 2281 |
+
"step": 1600
|
| 2282 |
+
},
|
| 2283 |
+
{
|
| 2284 |
+
"epoch": 2.56,
|
| 2285 |
+
"eval_chemistry_entropy": 1.1931069812774657,
|
| 2286 |
+
"eval_chemistry_loss": 1.278794527053833,
|
| 2287 |
+
"eval_chemistry_mean_token_accuracy": 0.7340463919639587,
|
| 2288 |
+
"eval_chemistry_num_tokens": 52964035.0,
|
| 2289 |
+
"eval_chemistry_runtime": 71.3365,
|
| 2290 |
+
"eval_chemistry_samples_per_second": 7.009,
|
| 2291 |
+
"eval_chemistry_steps_per_second": 1.752,
|
| 2292 |
+
"step": 1600
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.56,
|
| 2296 |
+
"eval_math_entropy": 0.8954266090393066,
|
| 2297 |
+
"eval_math_loss": 1.1417946815490723,
|
| 2298 |
+
"eval_math_mean_token_accuracy": 0.7577371735572815,
|
| 2299 |
+
"eval_math_num_tokens": 52964035.0,
|
| 2300 |
+
"eval_math_runtime": 74.3663,
|
| 2301 |
+
"eval_math_samples_per_second": 6.723,
|
| 2302 |
+
"eval_math_steps_per_second": 1.681,
|
| 2303 |
+
"step": 1600
|
| 2304 |
+
},
|
| 2305 |
+
{
|
| 2306 |
+
"epoch": 2.56,
|
| 2307 |
+
"eval_physics_entropy": 0.8706050214767456,
|
| 2308 |
+
"eval_physics_loss": 0.9212888479232788,
|
| 2309 |
+
"eval_physics_mean_token_accuracy": 0.7886158428192138,
|
| 2310 |
+
"eval_physics_num_tokens": 52964035.0,
|
| 2311 |
+
"eval_physics_runtime": 85.0105,
|
| 2312 |
+
"eval_physics_samples_per_second": 5.882,
|
| 2313 |
+
"eval_physics_steps_per_second": 1.47,
|
| 2314 |
+
"step": 1600
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"entropy": 0.8518974561244249,
|
| 2318 |
+
"epoch": 2.576,
|
| 2319 |
+
"grad_norm": 12.8125,
|
| 2320 |
+
"learning_rate": 1.0303703703703705e-05,
|
| 2321 |
+
"loss": 14.1697,
|
| 2322 |
+
"mean_token_accuracy": 0.7962028808891773,
|
| 2323 |
+
"num_tokens": 53293345.0,
|
| 2324 |
+
"step": 1610
|
| 2325 |
+
},
|
| 2326 |
+
{
|
| 2327 |
+
"entropy": 0.8667042430490255,
|
| 2328 |
+
"epoch": 2.592,
|
| 2329 |
+
"grad_norm": 10.625,
|
| 2330 |
+
"learning_rate": 1.0229629629629631e-05,
|
| 2331 |
+
"loss": 14.4307,
|
| 2332 |
+
"mean_token_accuracy": 0.7936250735074282,
|
| 2333 |
+
"num_tokens": 53620896.0,
|
| 2334 |
+
"step": 1620
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"entropy": 0.8402661351487041,
|
| 2338 |
+
"epoch": 2.608,
|
| 2339 |
+
"grad_norm": 11.25,
|
| 2340 |
+
"learning_rate": 1.0155555555555556e-05,
|
| 2341 |
+
"loss": 14.021,
|
| 2342 |
+
"mean_token_accuracy": 0.7978927873075008,
|
| 2343 |
+
"num_tokens": 53950989.0,
|
| 2344 |
+
"step": 1630
|
| 2345 |
+
},
|
| 2346 |
+
{
|
| 2347 |
+
"entropy": 0.8603313663974405,
|
| 2348 |
+
"epoch": 2.624,
|
| 2349 |
+
"grad_norm": 11.125,
|
| 2350 |
+
"learning_rate": 1.0081481481481484e-05,
|
| 2351 |
+
"loss": 14.3819,
|
| 2352 |
+
"mean_token_accuracy": 0.7938243903219699,
|
| 2353 |
+
"num_tokens": 54282732.0,
|
| 2354 |
+
"step": 1640
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"entropy": 0.8600944321602583,
|
| 2358 |
+
"epoch": 2.64,
|
| 2359 |
+
"grad_norm": 12.8125,
|
| 2360 |
+
"learning_rate": 1.0007407407407407e-05,
|
| 2361 |
+
"loss": 14.2866,
|
| 2362 |
+
"mean_token_accuracy": 0.7947955295443535,
|
| 2363 |
+
"num_tokens": 54609748.0,
|
| 2364 |
+
"step": 1650
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"entropy": 0.8892827957868576,
|
| 2368 |
+
"epoch": 2.656,
|
| 2369 |
+
"grad_norm": 12.5625,
|
| 2370 |
+
"learning_rate": 9.933333333333334e-06,
|
| 2371 |
+
"loss": 14.8315,
|
| 2372 |
+
"mean_token_accuracy": 0.7884532894939185,
|
| 2373 |
+
"num_tokens": 54953583.0,
|
| 2374 |
+
"step": 1660
|
| 2375 |
+
},
|
| 2376 |
+
{
|
| 2377 |
+
"entropy": 0.867825073376298,
|
| 2378 |
+
"epoch": 2.672,
|
| 2379 |
+
"grad_norm": 12.0,
|
| 2380 |
+
"learning_rate": 9.85925925925926e-06,
|
| 2381 |
+
"loss": 14.4804,
|
| 2382 |
+
"mean_token_accuracy": 0.791333881765604,
|
| 2383 |
+
"num_tokens": 55281945.0,
|
| 2384 |
+
"step": 1670
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"entropy": 0.8512159805744887,
|
| 2388 |
+
"epoch": 2.6879999999999997,
|
| 2389 |
+
"grad_norm": 12.0625,
|
| 2390 |
+
"learning_rate": 9.785185185185187e-06,
|
| 2391 |
+
"loss": 14.0027,
|
| 2392 |
+
"mean_token_accuracy": 0.798706853762269,
|
| 2393 |
+
"num_tokens": 55612391.0,
|
| 2394 |
+
"step": 1680
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"entropy": 0.8470454571768642,
|
| 2398 |
+
"epoch": 2.7039999999999997,
|
| 2399 |
+
"grad_norm": 11.25,
|
| 2400 |
+
"learning_rate": 9.711111111111111e-06,
|
| 2401 |
+
"loss": 14.2474,
|
| 2402 |
+
"mean_token_accuracy": 0.7953305941075086,
|
| 2403 |
+
"num_tokens": 55949791.0,
|
| 2404 |
+
"step": 1690
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"entropy": 0.8569748856127262,
|
| 2408 |
+
"epoch": 2.7199999999999998,
|
| 2409 |
+
"grad_norm": 14.625,
|
| 2410 |
+
"learning_rate": 9.637037037037038e-06,
|
| 2411 |
+
"loss": 14.2518,
|
| 2412 |
+
"mean_token_accuracy": 0.7947139970958232,
|
| 2413 |
+
"num_tokens": 56282204.0,
|
| 2414 |
+
"step": 1700
|
| 2415 |
+
},
|
| 2416 |
+
{
|
| 2417 |
+
"epoch": 2.7199999999999998,
|
| 2418 |
+
"eval_biology_entropy": 2.4623048429489134,
|
| 2419 |
+
"eval_biology_loss": 2.855733633041382,
|
| 2420 |
+
"eval_biology_mean_token_accuracy": 0.5338225281238556,
|
| 2421 |
+
"eval_biology_num_tokens": 56282204.0,
|
| 2422 |
+
"eval_biology_runtime": 57.2989,
|
| 2423 |
+
"eval_biology_samples_per_second": 8.726,
|
| 2424 |
+
"eval_biology_steps_per_second": 2.182,
|
| 2425 |
+
"step": 1700
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.7199999999999998,
|
| 2429 |
+
"eval_chemistry_entropy": 1.1981362085342406,
|
| 2430 |
+
"eval_chemistry_loss": 1.2777812480926514,
|
| 2431 |
+
"eval_chemistry_mean_token_accuracy": 0.7337445435523987,
|
| 2432 |
+
"eval_chemistry_num_tokens": 56282204.0,
|
| 2433 |
+
"eval_chemistry_runtime": 72.002,
|
| 2434 |
+
"eval_chemistry_samples_per_second": 6.944,
|
| 2435 |
+
"eval_chemistry_steps_per_second": 1.736,
|
| 2436 |
+
"step": 1700
|
| 2437 |
+
},
|
| 2438 |
+
{
|
| 2439 |
+
"epoch": 2.7199999999999998,
|
| 2440 |
+
"eval_math_entropy": 0.8990244059562683,
|
| 2441 |
+
"eval_math_loss": 1.1414705514907837,
|
| 2442 |
+
"eval_math_mean_token_accuracy": 0.7577006416320801,
|
| 2443 |
+
"eval_math_num_tokens": 56282204.0,
|
| 2444 |
+
"eval_math_runtime": 74.1316,
|
| 2445 |
+
"eval_math_samples_per_second": 6.745,
|
| 2446 |
+
"eval_math_steps_per_second": 1.686,
|
| 2447 |
+
"step": 1700
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.7199999999999998,
|
| 2451 |
+
"eval_physics_entropy": 0.8712537698745727,
|
| 2452 |
+
"eval_physics_loss": 0.9199900031089783,
|
| 2453 |
+
"eval_physics_mean_token_accuracy": 0.7888414950370789,
|
| 2454 |
+
"eval_physics_num_tokens": 56282204.0,
|
| 2455 |
+
"eval_physics_runtime": 85.5277,
|
| 2456 |
+
"eval_physics_samples_per_second": 5.846,
|
| 2457 |
+
"eval_physics_steps_per_second": 1.462,
|
| 2458 |
+
"step": 1700
|
| 2459 |
+
},
|
| 2460 |
+
{
|
| 2461 |
+
"entropy": 0.8889217928051949,
|
| 2462 |
+
"epoch": 2.7359999999999998,
|
| 2463 |
+
"grad_norm": 12.8125,
|
| 2464 |
+
"learning_rate": 9.562962962962965e-06,
|
| 2465 |
+
"loss": 14.8794,
|
| 2466 |
+
"mean_token_accuracy": 0.787835606187582,
|
| 2467 |
+
"num_tokens": 56605478.0,
|
| 2468 |
+
"step": 1710
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"entropy": 0.867193017527461,
|
| 2472 |
+
"epoch": 2.752,
|
| 2473 |
+
"grad_norm": 11.6875,
|
| 2474 |
+
"learning_rate": 9.48888888888889e-06,
|
| 2475 |
+
"loss": 14.5299,
|
| 2476 |
+
"mean_token_accuracy": 0.7921088628470898,
|
| 2477 |
+
"num_tokens": 56935874.0,
|
| 2478 |
+
"step": 1720
|
| 2479 |
+
},
|
| 2480 |
+
{
|
| 2481 |
+
"entropy": 0.8495670940726996,
|
| 2482 |
+
"epoch": 2.768,
|
| 2483 |
+
"grad_norm": 12.0,
|
| 2484 |
+
"learning_rate": 9.414814814814816e-06,
|
| 2485 |
+
"loss": 14.0654,
|
| 2486 |
+
"mean_token_accuracy": 0.7975914411246776,
|
| 2487 |
+
"num_tokens": 57270552.0,
|
| 2488 |
+
"step": 1730
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"entropy": 0.8686243133619428,
|
| 2492 |
+
"epoch": 2.784,
|
| 2493 |
+
"grad_norm": 12.5,
|
| 2494 |
+
"learning_rate": 9.34074074074074e-06,
|
| 2495 |
+
"loss": 14.4341,
|
| 2496 |
+
"mean_token_accuracy": 0.7923915140330792,
|
| 2497 |
+
"num_tokens": 57602968.0,
|
| 2498 |
+
"step": 1740
|
| 2499 |
+
},
|
| 2500 |
+
{
|
| 2501 |
+
"entropy": 0.9133183425292373,
|
| 2502 |
+
"epoch": 2.8,
|
| 2503 |
+
"grad_norm": 13.375,
|
| 2504 |
+
"learning_rate": 9.266666666666667e-06,
|
| 2505 |
+
"loss": 15.2497,
|
| 2506 |
+
"mean_token_accuracy": 0.7833729684352875,
|
| 2507 |
+
"num_tokens": 57925886.0,
|
| 2508 |
+
"step": 1750
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"entropy": 0.8448837269097567,
|
| 2512 |
+
"epoch": 2.816,
|
| 2513 |
+
"grad_norm": 11.875,
|
| 2514 |
+
"learning_rate": 9.192592592592594e-06,
|
| 2515 |
+
"loss": 13.9445,
|
| 2516 |
+
"mean_token_accuracy": 0.7979842260479927,
|
| 2517 |
+
"num_tokens": 58262960.0,
|
| 2518 |
+
"step": 1760
|
| 2519 |
+
},
|
| 2520 |
+
{
|
| 2521 |
+
"entropy": 0.871445388160646,
|
| 2522 |
+
"epoch": 2.832,
|
| 2523 |
+
"grad_norm": 12.3125,
|
| 2524 |
+
"learning_rate": 9.118518518518518e-06,
|
| 2525 |
+
"loss": 14.5598,
|
| 2526 |
+
"mean_token_accuracy": 0.7907900519669055,
|
| 2527 |
+
"num_tokens": 58586324.0,
|
| 2528 |
+
"step": 1770
|
| 2529 |
+
},
|
| 2530 |
+
{
|
| 2531 |
+
"entropy": 0.8576038857921958,
|
| 2532 |
+
"epoch": 2.848,
|
| 2533 |
+
"grad_norm": 13.0,
|
| 2534 |
+
"learning_rate": 9.044444444444445e-06,
|
| 2535 |
+
"loss": 14.3799,
|
| 2536 |
+
"mean_token_accuracy": 0.7930273406207562,
|
| 2537 |
+
"num_tokens": 58912949.0,
|
| 2538 |
+
"step": 1780
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"entropy": 0.8503576427698135,
|
| 2542 |
+
"epoch": 2.864,
|
| 2543 |
+
"grad_norm": 11.0,
|
| 2544 |
+
"learning_rate": 8.970370370370372e-06,
|
| 2545 |
+
"loss": 14.1151,
|
| 2546 |
+
"mean_token_accuracy": 0.7967234510928393,
|
| 2547 |
+
"num_tokens": 59252494.0,
|
| 2548 |
+
"step": 1790
|
| 2549 |
+
},
|
| 2550 |
+
{
|
| 2551 |
+
"entropy": 0.865095803141594,
|
| 2552 |
+
"epoch": 2.88,
|
| 2553 |
+
"grad_norm": 14.125,
|
| 2554 |
+
"learning_rate": 8.896296296296298e-06,
|
| 2555 |
+
"loss": 14.4,
|
| 2556 |
+
"mean_token_accuracy": 0.7936870157718658,
|
| 2557 |
+
"num_tokens": 59574951.0,
|
| 2558 |
+
"step": 1800
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 2.88,
|
| 2562 |
+
"eval_biology_entropy": 2.4605735635757444,
|
| 2563 |
+
"eval_biology_loss": 2.8438503742218018,
|
| 2564 |
+
"eval_biology_mean_token_accuracy": 0.5349292945861817,
|
| 2565 |
+
"eval_biology_num_tokens": 59574951.0,
|
| 2566 |
+
"eval_biology_runtime": 57.1516,
|
| 2567 |
+
"eval_biology_samples_per_second": 8.749,
|
| 2568 |
+
"eval_biology_steps_per_second": 2.187,
|
| 2569 |
+
"step": 1800
|
| 2570 |
+
},
|
| 2571 |
+
{
|
| 2572 |
+
"epoch": 2.88,
|
| 2573 |
+
"eval_chemistry_entropy": 1.1904199600219727,
|
| 2574 |
+
"eval_chemistry_loss": 1.2762060165405273,
|
| 2575 |
+
"eval_chemistry_mean_token_accuracy": 0.7342206130027771,
|
| 2576 |
+
"eval_chemistry_num_tokens": 59574951.0,
|
| 2577 |
+
"eval_chemistry_runtime": 71.4215,
|
| 2578 |
+
"eval_chemistry_samples_per_second": 7.001,
|
| 2579 |
+
"eval_chemistry_steps_per_second": 1.75,
|
| 2580 |
+
"step": 1800
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 2.88,
|
| 2584 |
+
"eval_math_entropy": 0.8948101491928101,
|
| 2585 |
+
"eval_math_loss": 1.1407443284988403,
|
| 2586 |
+
"eval_math_mean_token_accuracy": 0.7578761463165283,
|
| 2587 |
+
"eval_math_num_tokens": 59574951.0,
|
| 2588 |
+
"eval_math_runtime": 74.9351,
|
| 2589 |
+
"eval_math_samples_per_second": 6.672,
|
| 2590 |
+
"eval_math_steps_per_second": 1.668,
|
| 2591 |
+
"step": 1800
|
| 2592 |
+
},
|
| 2593 |
+
{
|
| 2594 |
+
"epoch": 2.88,
|
| 2595 |
+
"eval_physics_entropy": 0.8652872314453125,
|
| 2596 |
+
"eval_physics_loss": 0.9192015528678894,
|
| 2597 |
+
"eval_physics_mean_token_accuracy": 0.7890026559829711,
|
| 2598 |
+
"eval_physics_num_tokens": 59574951.0,
|
| 2599 |
+
"eval_physics_runtime": 84.6211,
|
| 2600 |
+
"eval_physics_samples_per_second": 5.909,
|
| 2601 |
+
"eval_physics_steps_per_second": 1.477,
|
| 2602 |
+
"step": 1800
|
| 2603 |
+
},
|
| 2604 |
+
{
|
| 2605 |
+
"entropy": 0.8504932520911097,
|
| 2606 |
+
"epoch": 2.896,
|
| 2607 |
+
"grad_norm": 12.75,
|
| 2608 |
+
"learning_rate": 8.822222222222223e-06,
|
| 2609 |
+
"loss": 14.2948,
|
| 2610 |
+
"mean_token_accuracy": 0.7946070168167353,
|
| 2611 |
+
"num_tokens": 59903274.0,
|
| 2612 |
+
"step": 1810
|
| 2613 |
+
},
|
| 2614 |
+
{
|
| 2615 |
+
"entropy": 0.8513661539182067,
|
| 2616 |
+
"epoch": 2.912,
|
| 2617 |
+
"grad_norm": 12.5625,
|
| 2618 |
+
"learning_rate": 8.74814814814815e-06,
|
| 2619 |
+
"loss": 14.1948,
|
| 2620 |
+
"mean_token_accuracy": 0.7956477370113134,
|
| 2621 |
+
"num_tokens": 60234203.0,
|
| 2622 |
+
"step": 1820
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"entropy": 0.8567061256617308,
|
| 2626 |
+
"epoch": 2.928,
|
| 2627 |
+
"grad_norm": 12.0625,
|
| 2628 |
+
"learning_rate": 8.674074074074074e-06,
|
| 2629 |
+
"loss": 14.2065,
|
| 2630 |
+
"mean_token_accuracy": 0.7965482659637928,
|
| 2631 |
+
"num_tokens": 60559123.0,
|
| 2632 |
+
"step": 1830
|
| 2633 |
+
},
|
| 2634 |
+
{
|
| 2635 |
+
"entropy": 0.8522207867354155,
|
| 2636 |
+
"epoch": 2.944,
|
| 2637 |
+
"grad_norm": 13.8125,
|
| 2638 |
+
"learning_rate": 8.6e-06,
|
| 2639 |
+
"loss": 14.1203,
|
| 2640 |
+
"mean_token_accuracy": 0.7966358475387096,
|
| 2641 |
+
"num_tokens": 60893423.0,
|
| 2642 |
+
"step": 1840
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"entropy": 0.8581688396632672,
|
| 2646 |
+
"epoch": 2.96,
|
| 2647 |
+
"grad_norm": 11.875,
|
| 2648 |
+
"learning_rate": 8.525925925925927e-06,
|
| 2649 |
+
"loss": 14.4079,
|
| 2650 |
+
"mean_token_accuracy": 0.7932413015514612,
|
| 2651 |
+
"num_tokens": 61228692.0,
|
| 2652 |
+
"step": 1850
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"entropy": 0.8690817683935166,
|
| 2656 |
+
"epoch": 2.976,
|
| 2657 |
+
"grad_norm": 11.9375,
|
| 2658 |
+
"learning_rate": 8.451851851851852e-06,
|
| 2659 |
+
"loss": 14.3962,
|
| 2660 |
+
"mean_token_accuracy": 0.793168543279171,
|
| 2661 |
+
"num_tokens": 61560852.0,
|
| 2662 |
+
"step": 1860
|
| 2663 |
+
},
|
| 2664 |
+
{
|
| 2665 |
+
"entropy": 0.8346356390044093,
|
| 2666 |
+
"epoch": 2.992,
|
| 2667 |
+
"grad_norm": 11.9375,
|
| 2668 |
+
"learning_rate": 8.377777777777779e-06,
|
| 2669 |
+
"loss": 13.939,
|
| 2670 |
+
"mean_token_accuracy": 0.7977076359093189,
|
| 2671 |
+
"num_tokens": 61898951.0,
|
| 2672 |
+
"step": 1870
|
| 2673 |
+
},
|
| 2674 |
+
{
|
| 2675 |
+
"entropy": 0.8470860140398144,
|
| 2676 |
+
"epoch": 3.008,
|
| 2677 |
+
"grad_norm": 11.4375,
|
| 2678 |
+
"learning_rate": 8.303703703703705e-06,
|
| 2679 |
+
"loss": 14.0544,
|
| 2680 |
+
"mean_token_accuracy": 0.7982985734939575,
|
| 2681 |
+
"num_tokens": 62223875.0,
|
| 2682 |
+
"step": 1880
|
| 2683 |
+
},
|
| 2684 |
+
{
|
| 2685 |
+
"entropy": 0.8579535936936736,
|
| 2686 |
+
"epoch": 3.024,
|
| 2687 |
+
"grad_norm": 12.0625,
|
| 2688 |
+
"learning_rate": 8.229629629629632e-06,
|
| 2689 |
+
"loss": 14.2346,
|
| 2690 |
+
"mean_token_accuracy": 0.7953595809638501,
|
| 2691 |
+
"num_tokens": 62553166.0,
|
| 2692 |
+
"step": 1890
|
| 2693 |
+
},
|
| 2694 |
+
{
|
| 2695 |
+
"entropy": 0.8405753003433347,
|
| 2696 |
+
"epoch": 3.04,
|
| 2697 |
+
"grad_norm": 12.0625,
|
| 2698 |
+
"learning_rate": 8.155555555555556e-06,
|
| 2699 |
+
"loss": 13.9614,
|
| 2700 |
+
"mean_token_accuracy": 0.7975836973637342,
|
| 2701 |
+
"num_tokens": 62893113.0,
|
| 2702 |
+
"step": 1900
|
| 2703 |
+
},
|
| 2704 |
+
{
|
| 2705 |
+
"epoch": 3.04,
|
| 2706 |
+
"eval_biology_entropy": 2.466928297996521,
|
| 2707 |
+
"eval_biology_loss": 2.8451387882232666,
|
| 2708 |
+
"eval_biology_mean_token_accuracy": 0.5350441763401032,
|
| 2709 |
+
"eval_biology_num_tokens": 62893113.0,
|
| 2710 |
+
"eval_biology_runtime": 57.2798,
|
| 2711 |
+
"eval_biology_samples_per_second": 8.729,
|
| 2712 |
+
"eval_biology_steps_per_second": 2.182,
|
| 2713 |
+
"step": 1900
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 3.04,
|
| 2717 |
+
"eval_chemistry_entropy": 1.1924911136627196,
|
| 2718 |
+
"eval_chemistry_loss": 1.2755722999572754,
|
| 2719 |
+
"eval_chemistry_mean_token_accuracy": 0.734372947216034,
|
| 2720 |
+
"eval_chemistry_num_tokens": 62893113.0,
|
| 2721 |
+
"eval_chemistry_runtime": 71.12,
|
| 2722 |
+
"eval_chemistry_samples_per_second": 7.03,
|
| 2723 |
+
"eval_chemistry_steps_per_second": 1.758,
|
| 2724 |
+
"step": 1900
|
| 2725 |
+
},
|
| 2726 |
+
{
|
| 2727 |
+
"epoch": 3.04,
|
| 2728 |
+
"eval_math_entropy": 0.8945438508987427,
|
| 2729 |
+
"eval_math_loss": 1.1409516334533691,
|
| 2730 |
+
"eval_math_mean_token_accuracy": 0.7579452800750732,
|
| 2731 |
+
"eval_math_num_tokens": 62893113.0,
|
| 2732 |
+
"eval_math_runtime": 74.6477,
|
| 2733 |
+
"eval_math_samples_per_second": 6.698,
|
| 2734 |
+
"eval_math_steps_per_second": 1.675,
|
| 2735 |
+
"step": 1900
|
| 2736 |
+
},
|
| 2737 |
+
{
|
| 2738 |
+
"epoch": 3.04,
|
| 2739 |
+
"eval_physics_entropy": 0.8654837455749512,
|
| 2740 |
+
"eval_physics_loss": 0.9181889295578003,
|
| 2741 |
+
"eval_physics_mean_token_accuracy": 0.7893532915115357,
|
| 2742 |
+
"eval_physics_num_tokens": 62893113.0,
|
| 2743 |
+
"eval_physics_runtime": 85.3356,
|
| 2744 |
+
"eval_physics_samples_per_second": 5.859,
|
| 2745 |
+
"eval_physics_steps_per_second": 1.465,
|
| 2746 |
+
"step": 1900
|
| 2747 |
+
},
|
| 2748 |
+
{
|
| 2749 |
+
"entropy": 0.8652172423899174,
|
| 2750 |
+
"epoch": 3.056,
|
| 2751 |
+
"grad_norm": 12.375,
|
| 2752 |
+
"learning_rate": 8.081481481481483e-06,
|
| 2753 |
+
"loss": 14.4707,
|
| 2754 |
+
"mean_token_accuracy": 0.7926124636083841,
|
| 2755 |
+
"num_tokens": 63218088.0,
|
| 2756 |
+
"step": 1910
|
| 2757 |
+
},
|
| 2758 |
+
{
|
| 2759 |
+
"entropy": 0.8625176247209311,
|
| 2760 |
+
"epoch": 3.072,
|
| 2761 |
+
"grad_norm": 13.5,
|
| 2762 |
+
"learning_rate": 8.007407407407408e-06,
|
| 2763 |
+
"loss": 14.2819,
|
| 2764 |
+
"mean_token_accuracy": 0.7927148371934891,
|
| 2765 |
+
"num_tokens": 63551201.0,
|
| 2766 |
+
"step": 1920
|
| 2767 |
+
},
|
| 2768 |
+
{
|
| 2769 |
+
"entropy": 0.8709696922451258,
|
| 2770 |
+
"epoch": 3.088,
|
| 2771 |
+
"grad_norm": 13.25,
|
| 2772 |
+
"learning_rate": 7.933333333333334e-06,
|
| 2773 |
+
"loss": 14.535,
|
| 2774 |
+
"mean_token_accuracy": 0.7909463763236999,
|
| 2775 |
+
"num_tokens": 63876362.0,
|
| 2776 |
+
"step": 1930
|
| 2777 |
+
},
|
| 2778 |
+
{
|
| 2779 |
+
"entropy": 0.8365857848897577,
|
| 2780 |
+
"epoch": 3.104,
|
| 2781 |
+
"grad_norm": 12.6875,
|
| 2782 |
+
"learning_rate": 7.859259259259259e-06,
|
| 2783 |
+
"loss": 13.917,
|
| 2784 |
+
"mean_token_accuracy": 0.7985875517129898,
|
| 2785 |
+
"num_tokens": 64208171.0,
|
| 2786 |
+
"step": 1940
|
| 2787 |
+
},
|
| 2788 |
+
{
|
| 2789 |
+
"entropy": 0.8436792023479939,
|
| 2790 |
+
"epoch": 3.12,
|
| 2791 |
+
"grad_norm": 12.5,
|
| 2792 |
+
"learning_rate": 7.785185185185185e-06,
|
| 2793 |
+
"loss": 13.9247,
|
| 2794 |
+
"mean_token_accuracy": 0.7994871154427529,
|
| 2795 |
+
"num_tokens": 64544493.0,
|
| 2796 |
+
"step": 1950
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"entropy": 0.8724946033209562,
|
| 2800 |
+
"epoch": 3.136,
|
| 2801 |
+
"grad_norm": 12.875,
|
| 2802 |
+
"learning_rate": 7.711111111111112e-06,
|
| 2803 |
+
"loss": 14.5896,
|
| 2804 |
+
"mean_token_accuracy": 0.790136469900608,
|
| 2805 |
+
"num_tokens": 64872488.0,
|
| 2806 |
+
"step": 1960
|
| 2807 |
+
},
|
| 2808 |
+
{
|
| 2809 |
+
"entropy": 0.8340439153835177,
|
| 2810 |
+
"epoch": 3.152,
|
| 2811 |
+
"grad_norm": 12.6875,
|
| 2812 |
+
"learning_rate": 7.637037037037037e-06,
|
| 2813 |
+
"loss": 13.9309,
|
| 2814 |
+
"mean_token_accuracy": 0.7989891562610865,
|
| 2815 |
+
"num_tokens": 65201124.0,
|
| 2816 |
+
"step": 1970
|
| 2817 |
+
},
|
| 2818 |
+
{
|
| 2819 |
+
"entropy": 0.8577351365238428,
|
| 2820 |
+
"epoch": 3.168,
|
| 2821 |
+
"grad_norm": 11.125,
|
| 2822 |
+
"learning_rate": 7.562962962962963e-06,
|
| 2823 |
+
"loss": 14.332,
|
| 2824 |
+
"mean_token_accuracy": 0.7956987973302603,
|
| 2825 |
+
"num_tokens": 65525868.0,
|
| 2826 |
+
"step": 1980
|
| 2827 |
+
},
|
| 2828 |
+
{
|
| 2829 |
+
"entropy": 0.8472510496154427,
|
| 2830 |
+
"epoch": 3.184,
|
| 2831 |
+
"grad_norm": 12.75,
|
| 2832 |
+
"learning_rate": 7.48888888888889e-06,
|
| 2833 |
+
"loss": 14.0577,
|
| 2834 |
+
"mean_token_accuracy": 0.7991137109696865,
|
| 2835 |
+
"num_tokens": 65861603.0,
|
| 2836 |
+
"step": 1990
|
| 2837 |
+
},
|
| 2838 |
+
{
|
| 2839 |
+
"entropy": 0.8585745912045241,
|
| 2840 |
+
"epoch": 3.2,
|
| 2841 |
+
"grad_norm": 11.3125,
|
| 2842 |
+
"learning_rate": 7.4148148148148155e-06,
|
| 2843 |
+
"loss": 14.2601,
|
| 2844 |
+
"mean_token_accuracy": 0.7946750316768885,
|
| 2845 |
+
"num_tokens": 66193675.0,
|
| 2846 |
+
"step": 2000
|
| 2847 |
+
},
|
| 2848 |
+
{
|
| 2849 |
+
"epoch": 3.2,
|
| 2850 |
+
"eval_biology_entropy": 2.4529026355743406,
|
| 2851 |
+
"eval_biology_loss": 2.8515477180480957,
|
| 2852 |
+
"eval_biology_mean_token_accuracy": 0.5361672310829163,
|
| 2853 |
+
"eval_biology_num_tokens": 66193675.0,
|
| 2854 |
+
"eval_biology_runtime": 56.9907,
|
| 2855 |
+
"eval_biology_samples_per_second": 8.773,
|
| 2856 |
+
"eval_biology_steps_per_second": 2.193,
|
| 2857 |
+
"step": 2000
|
| 2858 |
+
},
|
| 2859 |
+
{
|
| 2860 |
+
"epoch": 3.2,
|
| 2861 |
+
"eval_chemistry_entropy": 1.1896215171813964,
|
| 2862 |
+
"eval_chemistry_loss": 1.2748806476593018,
|
| 2863 |
+
"eval_chemistry_mean_token_accuracy": 0.7352399125099182,
|
| 2864 |
+
"eval_chemistry_num_tokens": 66193675.0,
|
| 2865 |
+
"eval_chemistry_runtime": 71.1598,
|
| 2866 |
+
"eval_chemistry_samples_per_second": 7.026,
|
| 2867 |
+
"eval_chemistry_steps_per_second": 1.757,
|
| 2868 |
+
"step": 2000
|
| 2869 |
+
},
|
| 2870 |
+
{
|
| 2871 |
+
"epoch": 3.2,
|
| 2872 |
+
"eval_math_entropy": 0.8935150079727173,
|
| 2873 |
+
"eval_math_loss": 1.140594482421875,
|
| 2874 |
+
"eval_math_mean_token_accuracy": 0.7576922287940979,
|
| 2875 |
+
"eval_math_num_tokens": 66193675.0,
|
| 2876 |
+
"eval_math_runtime": 74.3826,
|
| 2877 |
+
"eval_math_samples_per_second": 6.722,
|
| 2878 |
+
"eval_math_steps_per_second": 1.681,
|
| 2879 |
+
"step": 2000
|
| 2880 |
+
},
|
| 2881 |
+
{
|
| 2882 |
+
"epoch": 3.2,
|
| 2883 |
+
"eval_physics_entropy": 0.8648786902427673,
|
| 2884 |
+
"eval_physics_loss": 0.9176769852638245,
|
| 2885 |
+
"eval_physics_mean_token_accuracy": 0.7893374056816101,
|
| 2886 |
+
"eval_physics_num_tokens": 66193675.0,
|
| 2887 |
+
"eval_physics_runtime": 84.6448,
|
| 2888 |
+
"eval_physics_samples_per_second": 5.907,
|
| 2889 |
+
"eval_physics_steps_per_second": 1.477,
|
| 2890 |
+
"step": 2000
|
| 2891 |
+
}
|
| 2892 |
+
],
|
| 2893 |
+
"logging_steps": 10,
|
| 2894 |
+
"max_steps": 3000,
|
| 2895 |
+
"num_input_tokens_seen": 0,
|
| 2896 |
+
"num_train_epochs": 5,
|
| 2897 |
+
"save_steps": 500,
|
| 2898 |
+
"stateful_callbacks": {
|
| 2899 |
+
"TrainerControl": {
|
| 2900 |
+
"args": {
|
| 2901 |
+
"should_epoch_stop": false,
|
| 2902 |
+
"should_evaluate": false,
|
| 2903 |
+
"should_log": false,
|
| 2904 |
+
"should_save": true,
|
| 2905 |
+
"should_training_stop": false
|
| 2906 |
+
},
|
| 2907 |
+
"attributes": {}
|
| 2908 |
+
}
|
| 2909 |
+
},
|
| 2910 |
+
"total_flos": 6.614072648666411e+18,
|
| 2911 |
+
"train_batch_size": 4,
|
| 2912 |
+
"trial_name": null,
|
| 2913 |
+
"trial_params": null
|
| 2914 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:931abe455d5684f7fdb033d2616f7c607fe28af43a0be9856fe4a9c238858acc
|
| 3 |
+
size 6417
|