Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- checkpoint-1000/added_tokens.json +3 -0
- checkpoint-1000/chat_template.jinja +47 -0
- checkpoint-1000/config.json +110 -0
- checkpoint-1000/generation_config.json +13 -0
- checkpoint-1000/model-00001-of-00005.safetensors +3 -0
- checkpoint-1000/model-00002-of-00005.safetensors +3 -0
- checkpoint-1000/model-00003-of-00005.safetensors +3 -0
- checkpoint-1000/model-00004-of-00005.safetensors +3 -0
- checkpoint-1000/model-00005-of-00005.safetensors +3 -0
- checkpoint-1000/model.safetensors.index.json +0 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/special_tokens_map.json +33 -0
- checkpoint-1000/tokenizer.json +3 -0
- checkpoint-1000/tokenizer.model +3 -0
- checkpoint-1000/tokenizer_config.json +0 -0
- checkpoint-1000/trainer_state.json +1254 -0
- checkpoint-1000/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -34,3 +34,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-1000/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image_soft_token>": 262144
|
| 3 |
+
}
|
checkpoint-1000/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-1000/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-1000/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-1000/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-1000/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-1000/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-1000/model-00004-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee311946fa78466bce3a329851413fe17f58b86a7f6e3f4f08bc55a00977adba
|
| 3 |
+
size 4931296656
|
checkpoint-1000/model-00005-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ac669c02a8193b4092dd25a161cce28fe258e7c3f02fb24584cbdfb45d48f49
|
| 3 |
+
size 4601000928
|
checkpoint-1000/model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:377d491e701fcd6978c3bce3bf1b6d2bffda858e1d5ace18d07606d9abf1f8cf
|
| 3 |
+
size 7286558019
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:acd74ad34e3c5060506b53a8a3fafbfb35d44065ee496b2abf3a0b9a3b93ba09
|
| 3 |
+
size 14645
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6078e52773a379f026b8fbbbc5546ab7ed9418d05b713f8aff2a4f3c7e12f108
|
| 3 |
+
size 1465
|
checkpoint-1000/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-1000/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
|
| 3 |
+
size 33384568
|
checkpoint-1000/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
|
| 3 |
+
size 4689074
|
checkpoint-1000/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,1254 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.4,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 0.7581225659698247,
|
| 14 |
+
"epoch": 0.004,
|
| 15 |
+
"grad_norm": 101.5,
|
| 16 |
+
"learning_rate": 1.8e-07,
|
| 17 |
+
"loss": 18.1141,
|
| 18 |
+
"mean_token_accuracy": 0.7393306585028767,
|
| 19 |
+
"num_tokens": 72542.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 0.7520309612154961,
|
| 24 |
+
"epoch": 0.008,
|
| 25 |
+
"grad_norm": 162.0,
|
| 26 |
+
"learning_rate": 3.8e-07,
|
| 27 |
+
"loss": 18.2441,
|
| 28 |
+
"mean_token_accuracy": 0.7392390985041857,
|
| 29 |
+
"num_tokens": 138146.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 0.741816989146173,
|
| 34 |
+
"epoch": 0.012,
|
| 35 |
+
"grad_norm": 92.5,
|
| 36 |
+
"learning_rate": 5.800000000000001e-07,
|
| 37 |
+
"loss": 17.8539,
|
| 38 |
+
"mean_token_accuracy": 0.7405363619327545,
|
| 39 |
+
"num_tokens": 208409.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 0.7636421315371991,
|
| 44 |
+
"epoch": 0.016,
|
| 45 |
+
"grad_norm": 88.0,
|
| 46 |
+
"learning_rate": 7.8e-07,
|
| 47 |
+
"loss": 18.142,
|
| 48 |
+
"mean_token_accuracy": 0.7400334309786558,
|
| 49 |
+
"num_tokens": 280941.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 0.7905357209965587,
|
| 54 |
+
"epoch": 0.02,
|
| 55 |
+
"grad_norm": 108.0,
|
| 56 |
+
"learning_rate": 9.800000000000001e-07,
|
| 57 |
+
"loss": 18.1041,
|
| 58 |
+
"mean_token_accuracy": 0.7354144401848316,
|
| 59 |
+
"num_tokens": 349731.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 0.8223027169704438,
|
| 64 |
+
"epoch": 0.024,
|
| 65 |
+
"grad_norm": 112.5,
|
| 66 |
+
"learning_rate": 1.1800000000000001e-06,
|
| 67 |
+
"loss": 18.4599,
|
| 68 |
+
"mean_token_accuracy": 0.731158659607172,
|
| 69 |
+
"num_tokens": 419127.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 0.7908559931442142,
|
| 74 |
+
"epoch": 0.028,
|
| 75 |
+
"grad_norm": 106.0,
|
| 76 |
+
"learning_rate": 1.3800000000000001e-06,
|
| 77 |
+
"loss": 17.5053,
|
| 78 |
+
"mean_token_accuracy": 0.7440328601747751,
|
| 79 |
+
"num_tokens": 489016.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 0.8229548260569572,
|
| 84 |
+
"epoch": 0.032,
|
| 85 |
+
"grad_norm": 86.0,
|
| 86 |
+
"learning_rate": 1.5800000000000001e-06,
|
| 87 |
+
"loss": 17.3714,
|
| 88 |
+
"mean_token_accuracy": 0.7438380591571331,
|
| 89 |
+
"num_tokens": 558056.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 0.84512098710984,
|
| 94 |
+
"epoch": 0.036,
|
| 95 |
+
"grad_norm": 100.5,
|
| 96 |
+
"learning_rate": 1.7800000000000001e-06,
|
| 97 |
+
"loss": 17.2456,
|
| 98 |
+
"mean_token_accuracy": 0.7478047309443354,
|
| 99 |
+
"num_tokens": 627804.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 0.8738276792690158,
|
| 104 |
+
"epoch": 0.04,
|
| 105 |
+
"grad_norm": 67.0,
|
| 106 |
+
"learning_rate": 1.98e-06,
|
| 107 |
+
"loss": 16.6896,
|
| 108 |
+
"mean_token_accuracy": 0.748936639726162,
|
| 109 |
+
"num_tokens": 697556.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.04,
|
| 114 |
+
"eval_biology_entropy": 1.1379984815716744,
|
| 115 |
+
"eval_biology_loss": 1.2692387104034424,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.6880569371581078,
|
| 117 |
+
"eval_biology_num_tokens": 697556.0,
|
| 118 |
+
"eval_biology_runtime": 40.5573,
|
| 119 |
+
"eval_biology_samples_per_second": 12.328,
|
| 120 |
+
"eval_biology_steps_per_second": 12.328,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.04,
|
| 125 |
+
"eval_math_entropy": 0.9095822716951371,
|
| 126 |
+
"eval_math_loss": 1.0899367332458496,
|
| 127 |
+
"eval_math_mean_token_accuracy": 0.7395683622956276,
|
| 128 |
+
"eval_math_num_tokens": 697556.0,
|
| 129 |
+
"eval_math_runtime": 48.0541,
|
| 130 |
+
"eval_math_samples_per_second": 10.405,
|
| 131 |
+
"eval_math_steps_per_second": 10.405,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"entropy": 0.9193563124164938,
|
| 136 |
+
"epoch": 0.044,
|
| 137 |
+
"grad_norm": 88.5,
|
| 138 |
+
"learning_rate": 2.1800000000000003e-06,
|
| 139 |
+
"loss": 17.3763,
|
| 140 |
+
"mean_token_accuracy": 0.7420263335108757,
|
| 141 |
+
"num_tokens": 765262.0,
|
| 142 |
+
"step": 110
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"entropy": 0.8906641118228436,
|
| 146 |
+
"epoch": 0.048,
|
| 147 |
+
"grad_norm": 78.0,
|
| 148 |
+
"learning_rate": 2.38e-06,
|
| 149 |
+
"loss": 16.2142,
|
| 150 |
+
"mean_token_accuracy": 0.7552934620529413,
|
| 151 |
+
"num_tokens": 836753.0,
|
| 152 |
+
"step": 120
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"entropy": 0.9178624864667654,
|
| 156 |
+
"epoch": 0.052,
|
| 157 |
+
"grad_norm": 70.0,
|
| 158 |
+
"learning_rate": 2.5800000000000003e-06,
|
| 159 |
+
"loss": 16.3594,
|
| 160 |
+
"mean_token_accuracy": 0.7507667759433388,
|
| 161 |
+
"num_tokens": 909947.0,
|
| 162 |
+
"step": 130
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"entropy": 0.9556385111063719,
|
| 166 |
+
"epoch": 0.056,
|
| 167 |
+
"grad_norm": 70.0,
|
| 168 |
+
"learning_rate": 2.7800000000000005e-06,
|
| 169 |
+
"loss": 16.6438,
|
| 170 |
+
"mean_token_accuracy": 0.748611630871892,
|
| 171 |
+
"num_tokens": 981959.0,
|
| 172 |
+
"step": 140
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"entropy": 0.9541467605158687,
|
| 176 |
+
"epoch": 0.06,
|
| 177 |
+
"grad_norm": 86.0,
|
| 178 |
+
"learning_rate": 2.9800000000000003e-06,
|
| 179 |
+
"loss": 16.5304,
|
| 180 |
+
"mean_token_accuracy": 0.7486465960741043,
|
| 181 |
+
"num_tokens": 1053624.0,
|
| 182 |
+
"step": 150
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"entropy": 0.9960285406559706,
|
| 186 |
+
"epoch": 0.064,
|
| 187 |
+
"grad_norm": 78.0,
|
| 188 |
+
"learning_rate": 3.1800000000000005e-06,
|
| 189 |
+
"loss": 16.9435,
|
| 190 |
+
"mean_token_accuracy": 0.7448480188846588,
|
| 191 |
+
"num_tokens": 1126446.0,
|
| 192 |
+
"step": 160
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"entropy": 0.9677659103646874,
|
| 196 |
+
"epoch": 0.068,
|
| 197 |
+
"grad_norm": 106.5,
|
| 198 |
+
"learning_rate": 3.3800000000000007e-06,
|
| 199 |
+
"loss": 16.1013,
|
| 200 |
+
"mean_token_accuracy": 0.7561902595683933,
|
| 201 |
+
"num_tokens": 1196286.0,
|
| 202 |
+
"step": 170
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"entropy": 0.9270429328083992,
|
| 206 |
+
"epoch": 0.072,
|
| 207 |
+
"grad_norm": 62.0,
|
| 208 |
+
"learning_rate": 3.58e-06,
|
| 209 |
+
"loss": 15.3741,
|
| 210 |
+
"mean_token_accuracy": 0.763100459985435,
|
| 211 |
+
"num_tokens": 1273905.0,
|
| 212 |
+
"step": 180
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"entropy": 0.9678281666710973,
|
| 216 |
+
"epoch": 0.076,
|
| 217 |
+
"grad_norm": 51.0,
|
| 218 |
+
"learning_rate": 3.7800000000000002e-06,
|
| 219 |
+
"loss": 15.8618,
|
| 220 |
+
"mean_token_accuracy": 0.7534618094563484,
|
| 221 |
+
"num_tokens": 1346454.0,
|
| 222 |
+
"step": 190
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"entropy": 0.9695806819945574,
|
| 226 |
+
"epoch": 0.08,
|
| 227 |
+
"grad_norm": 61.25,
|
| 228 |
+
"learning_rate": 3.980000000000001e-06,
|
| 229 |
+
"loss": 15.7874,
|
| 230 |
+
"mean_token_accuracy": 0.7560878146439791,
|
| 231 |
+
"num_tokens": 1413596.0,
|
| 232 |
+
"step": 200
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.08,
|
| 236 |
+
"eval_biology_entropy": 1.214613477051258,
|
| 237 |
+
"eval_biology_loss": 1.2630934715270996,
|
| 238 |
+
"eval_biology_mean_token_accuracy": 0.6885634814500808,
|
| 239 |
+
"eval_biology_num_tokens": 1413596.0,
|
| 240 |
+
"eval_biology_runtime": 40.4294,
|
| 241 |
+
"eval_biology_samples_per_second": 12.367,
|
| 242 |
+
"eval_biology_steps_per_second": 12.367,
|
| 243 |
+
"step": 200
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.08,
|
| 247 |
+
"eval_math_entropy": 0.9574190213680267,
|
| 248 |
+
"eval_math_loss": 0.9871246218681335,
|
| 249 |
+
"eval_math_mean_token_accuracy": 0.7557044946551323,
|
| 250 |
+
"eval_math_num_tokens": 1413596.0,
|
| 251 |
+
"eval_math_runtime": 48.0895,
|
| 252 |
+
"eval_math_samples_per_second": 10.397,
|
| 253 |
+
"eval_math_steps_per_second": 10.397,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"entropy": 0.8888338362798095,
|
| 258 |
+
"epoch": 0.084,
|
| 259 |
+
"grad_norm": 61.25,
|
| 260 |
+
"learning_rate": 4.18e-06,
|
| 261 |
+
"loss": 14.623,
|
| 262 |
+
"mean_token_accuracy": 0.7706882227212191,
|
| 263 |
+
"num_tokens": 1485147.0,
|
| 264 |
+
"step": 210
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"entropy": 0.906145141646266,
|
| 268 |
+
"epoch": 0.088,
|
| 269 |
+
"grad_norm": 53.5,
|
| 270 |
+
"learning_rate": 4.38e-06,
|
| 271 |
+
"loss": 14.8128,
|
| 272 |
+
"mean_token_accuracy": 0.7661576565355063,
|
| 273 |
+
"num_tokens": 1559450.0,
|
| 274 |
+
"step": 220
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"entropy": 0.95110107883811,
|
| 278 |
+
"epoch": 0.092,
|
| 279 |
+
"grad_norm": 63.25,
|
| 280 |
+
"learning_rate": 4.58e-06,
|
| 281 |
+
"loss": 15.2457,
|
| 282 |
+
"mean_token_accuracy": 0.761219710484147,
|
| 283 |
+
"num_tokens": 1631535.0,
|
| 284 |
+
"step": 230
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"entropy": 0.9411580273881555,
|
| 288 |
+
"epoch": 0.096,
|
| 289 |
+
"grad_norm": 51.0,
|
| 290 |
+
"learning_rate": 4.78e-06,
|
| 291 |
+
"loss": 14.938,
|
| 292 |
+
"mean_token_accuracy": 0.7646324578672647,
|
| 293 |
+
"num_tokens": 1701193.0,
|
| 294 |
+
"step": 240
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"entropy": 0.9254088591784239,
|
| 298 |
+
"epoch": 0.1,
|
| 299 |
+
"grad_norm": 56.25,
|
| 300 |
+
"learning_rate": 4.980000000000001e-06,
|
| 301 |
+
"loss": 14.6029,
|
| 302 |
+
"mean_token_accuracy": 0.7678503680974245,
|
| 303 |
+
"num_tokens": 1773596.0,
|
| 304 |
+
"step": 250
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"entropy": 0.9214777417480946,
|
| 308 |
+
"epoch": 0.104,
|
| 309 |
+
"grad_norm": 54.0,
|
| 310 |
+
"learning_rate": 5.18e-06,
|
| 311 |
+
"loss": 15.043,
|
| 312 |
+
"mean_token_accuracy": 0.7618352189660073,
|
| 313 |
+
"num_tokens": 1840165.0,
|
| 314 |
+
"step": 260
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"entropy": 0.9363527970388532,
|
| 318 |
+
"epoch": 0.108,
|
| 319 |
+
"grad_norm": 111.5,
|
| 320 |
+
"learning_rate": 5.380000000000001e-06,
|
| 321 |
+
"loss": 15.6052,
|
| 322 |
+
"mean_token_accuracy": 0.7564919672906398,
|
| 323 |
+
"num_tokens": 1908149.0,
|
| 324 |
+
"step": 270
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"entropy": 0.9113753678277134,
|
| 328 |
+
"epoch": 0.112,
|
| 329 |
+
"grad_norm": 53.0,
|
| 330 |
+
"learning_rate": 5.580000000000001e-06,
|
| 331 |
+
"loss": 14.4645,
|
| 332 |
+
"mean_token_accuracy": 0.7736385434865951,
|
| 333 |
+
"num_tokens": 1979232.0,
|
| 334 |
+
"step": 280
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"entropy": 0.916975767724216,
|
| 338 |
+
"epoch": 0.116,
|
| 339 |
+
"grad_norm": 54.5,
|
| 340 |
+
"learning_rate": 5.78e-06,
|
| 341 |
+
"loss": 15.0318,
|
| 342 |
+
"mean_token_accuracy": 0.7656373519450426,
|
| 343 |
+
"num_tokens": 2052039.0,
|
| 344 |
+
"step": 290
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"entropy": 0.915022067911923,
|
| 348 |
+
"epoch": 0.12,
|
| 349 |
+
"grad_norm": 46.25,
|
| 350 |
+
"learning_rate": 5.98e-06,
|
| 351 |
+
"loss": 14.7236,
|
| 352 |
+
"mean_token_accuracy": 0.7691222932189703,
|
| 353 |
+
"num_tokens": 2125815.0,
|
| 354 |
+
"step": 300
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.12,
|
| 358 |
+
"eval_biology_entropy": 1.2286294117569923,
|
| 359 |
+
"eval_biology_loss": 1.268429160118103,
|
| 360 |
+
"eval_biology_mean_token_accuracy": 0.6861296361088752,
|
| 361 |
+
"eval_biology_num_tokens": 2125815.0,
|
| 362 |
+
"eval_biology_runtime": 40.4544,
|
| 363 |
+
"eval_biology_samples_per_second": 12.36,
|
| 364 |
+
"eval_biology_steps_per_second": 12.36,
|
| 365 |
+
"step": 300
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 0.12,
|
| 369 |
+
"eval_math_entropy": 0.9302992832064628,
|
| 370 |
+
"eval_math_loss": 0.9273415803909302,
|
| 371 |
+
"eval_math_mean_token_accuracy": 0.7664831226468086,
|
| 372 |
+
"eval_math_num_tokens": 2125815.0,
|
| 373 |
+
"eval_math_runtime": 48.047,
|
| 374 |
+
"eval_math_samples_per_second": 10.406,
|
| 375 |
+
"eval_math_steps_per_second": 10.406,
|
| 376 |
+
"step": 300
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"entropy": 0.9336849415674806,
|
| 380 |
+
"epoch": 0.124,
|
| 381 |
+
"grad_norm": 55.25,
|
| 382 |
+
"learning_rate": 6.18e-06,
|
| 383 |
+
"loss": 14.9674,
|
| 384 |
+
"mean_token_accuracy": 0.7661335166543722,
|
| 385 |
+
"num_tokens": 2201991.0,
|
| 386 |
+
"step": 310
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"entropy": 0.8756070390343667,
|
| 390 |
+
"epoch": 0.128,
|
| 391 |
+
"grad_norm": 52.25,
|
| 392 |
+
"learning_rate": 6.380000000000001e-06,
|
| 393 |
+
"loss": 14.0,
|
| 394 |
+
"mean_token_accuracy": 0.7786776568740607,
|
| 395 |
+
"num_tokens": 2274177.0,
|
| 396 |
+
"step": 320
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"entropy": 0.892756593041122,
|
| 400 |
+
"epoch": 0.132,
|
| 401 |
+
"grad_norm": 45.75,
|
| 402 |
+
"learning_rate": 6.5800000000000005e-06,
|
| 403 |
+
"loss": 14.1896,
|
| 404 |
+
"mean_token_accuracy": 0.7731969062238931,
|
| 405 |
+
"num_tokens": 2342173.0,
|
| 406 |
+
"step": 330
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"entropy": 0.8899674592539668,
|
| 410 |
+
"epoch": 0.136,
|
| 411 |
+
"grad_norm": 56.5,
|
| 412 |
+
"learning_rate": 6.780000000000001e-06,
|
| 413 |
+
"loss": 14.2167,
|
| 414 |
+
"mean_token_accuracy": 0.7683363553136587,
|
| 415 |
+
"num_tokens": 2412435.0,
|
| 416 |
+
"step": 340
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"entropy": 0.9237651886418462,
|
| 420 |
+
"epoch": 0.14,
|
| 421 |
+
"grad_norm": 49.5,
|
| 422 |
+
"learning_rate": 6.98e-06,
|
| 423 |
+
"loss": 14.7203,
|
| 424 |
+
"mean_token_accuracy": 0.7684664275497198,
|
| 425 |
+
"num_tokens": 2481105.0,
|
| 426 |
+
"step": 350
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"entropy": 0.8494639024138451,
|
| 430 |
+
"epoch": 0.144,
|
| 431 |
+
"grad_norm": 49.0,
|
| 432 |
+
"learning_rate": 7.180000000000001e-06,
|
| 433 |
+
"loss": 13.6956,
|
| 434 |
+
"mean_token_accuracy": 0.7817885402590037,
|
| 435 |
+
"num_tokens": 2548445.0,
|
| 436 |
+
"step": 360
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"entropy": 0.8341110894456506,
|
| 440 |
+
"epoch": 0.148,
|
| 441 |
+
"grad_norm": 44.0,
|
| 442 |
+
"learning_rate": 7.3800000000000005e-06,
|
| 443 |
+
"loss": 13.2223,
|
| 444 |
+
"mean_token_accuracy": 0.789648124948144,
|
| 445 |
+
"num_tokens": 2621364.0,
|
| 446 |
+
"step": 370
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"entropy": 0.8608193326741457,
|
| 450 |
+
"epoch": 0.152,
|
| 451 |
+
"grad_norm": 52.25,
|
| 452 |
+
"learning_rate": 7.58e-06,
|
| 453 |
+
"loss": 13.8861,
|
| 454 |
+
"mean_token_accuracy": 0.7754677496850491,
|
| 455 |
+
"num_tokens": 2686462.0,
|
| 456 |
+
"step": 380
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"entropy": 0.8447565982118249,
|
| 460 |
+
"epoch": 0.156,
|
| 461 |
+
"grad_norm": 50.0,
|
| 462 |
+
"learning_rate": 7.78e-06,
|
| 463 |
+
"loss": 13.5199,
|
| 464 |
+
"mean_token_accuracy": 0.7812942665070295,
|
| 465 |
+
"num_tokens": 2754241.0,
|
| 466 |
+
"step": 390
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"entropy": 0.8711243279278278,
|
| 470 |
+
"epoch": 0.16,
|
| 471 |
+
"grad_norm": 79.0,
|
| 472 |
+
"learning_rate": 7.980000000000002e-06,
|
| 473 |
+
"loss": 14.1276,
|
| 474 |
+
"mean_token_accuracy": 0.7742310840636492,
|
| 475 |
+
"num_tokens": 2824418.0,
|
| 476 |
+
"step": 400
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.16,
|
| 480 |
+
"eval_biology_entropy": 1.22758097666502,
|
| 481 |
+
"eval_biology_loss": 1.2762199640274048,
|
| 482 |
+
"eval_biology_mean_token_accuracy": 0.6836535354852676,
|
| 483 |
+
"eval_biology_num_tokens": 2824418.0,
|
| 484 |
+
"eval_biology_runtime": 40.7281,
|
| 485 |
+
"eval_biology_samples_per_second": 12.277,
|
| 486 |
+
"eval_biology_steps_per_second": 12.277,
|
| 487 |
+
"step": 400
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 0.16,
|
| 491 |
+
"eval_math_entropy": 0.9047691844701767,
|
| 492 |
+
"eval_math_loss": 0.8841754198074341,
|
| 493 |
+
"eval_math_mean_token_accuracy": 0.775034920334816,
|
| 494 |
+
"eval_math_num_tokens": 2824418.0,
|
| 495 |
+
"eval_math_runtime": 47.9197,
|
| 496 |
+
"eval_math_samples_per_second": 10.434,
|
| 497 |
+
"eval_math_steps_per_second": 10.434,
|
| 498 |
+
"step": 400
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"entropy": 0.8859184484928846,
|
| 502 |
+
"epoch": 0.164,
|
| 503 |
+
"grad_norm": 127.0,
|
| 504 |
+
"learning_rate": 8.18e-06,
|
| 505 |
+
"loss": 14.142,
|
| 506 |
+
"mean_token_accuracy": 0.7765266574919224,
|
| 507 |
+
"num_tokens": 2895667.0,
|
| 508 |
+
"step": 410
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"entropy": 0.8501209424808621,
|
| 512 |
+
"epoch": 0.168,
|
| 513 |
+
"grad_norm": 40.0,
|
| 514 |
+
"learning_rate": 8.380000000000001e-06,
|
| 515 |
+
"loss": 13.8435,
|
| 516 |
+
"mean_token_accuracy": 0.7774271417409182,
|
| 517 |
+
"num_tokens": 2966031.0,
|
| 518 |
+
"step": 420
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"entropy": 0.8510567891411484,
|
| 522 |
+
"epoch": 0.172,
|
| 523 |
+
"grad_norm": 45.75,
|
| 524 |
+
"learning_rate": 8.580000000000001e-06,
|
| 525 |
+
"loss": 13.3673,
|
| 526 |
+
"mean_token_accuracy": 0.7858892437070608,
|
| 527 |
+
"num_tokens": 3039780.0,
|
| 528 |
+
"step": 430
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"entropy": 0.8355857279151678,
|
| 532 |
+
"epoch": 0.176,
|
| 533 |
+
"grad_norm": 44.5,
|
| 534 |
+
"learning_rate": 8.78e-06,
|
| 535 |
+
"loss": 13.7383,
|
| 536 |
+
"mean_token_accuracy": 0.7784004233777523,
|
| 537 |
+
"num_tokens": 3110313.0,
|
| 538 |
+
"step": 440
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"entropy": 0.8662491450086236,
|
| 542 |
+
"epoch": 0.18,
|
| 543 |
+
"grad_norm": 43.75,
|
| 544 |
+
"learning_rate": 8.98e-06,
|
| 545 |
+
"loss": 13.6485,
|
| 546 |
+
"mean_token_accuracy": 0.7806661173701286,
|
| 547 |
+
"num_tokens": 3179963.0,
|
| 548 |
+
"step": 450
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"entropy": 0.8596813324838877,
|
| 552 |
+
"epoch": 0.184,
|
| 553 |
+
"grad_norm": 46.0,
|
| 554 |
+
"learning_rate": 9.180000000000002e-06,
|
| 555 |
+
"loss": 13.7717,
|
| 556 |
+
"mean_token_accuracy": 0.7796813258901238,
|
| 557 |
+
"num_tokens": 3251139.0,
|
| 558 |
+
"step": 460
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"entropy": 0.8454847941175103,
|
| 562 |
+
"epoch": 0.188,
|
| 563 |
+
"grad_norm": 56.25,
|
| 564 |
+
"learning_rate": 9.38e-06,
|
| 565 |
+
"loss": 13.4682,
|
| 566 |
+
"mean_token_accuracy": 0.7803339377045632,
|
| 567 |
+
"num_tokens": 3323519.0,
|
| 568 |
+
"step": 470
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"entropy": 0.8228314617648721,
|
| 572 |
+
"epoch": 0.192,
|
| 573 |
+
"grad_norm": 49.75,
|
| 574 |
+
"learning_rate": 9.58e-06,
|
| 575 |
+
"loss": 13.3842,
|
| 576 |
+
"mean_token_accuracy": 0.7827629748731851,
|
| 577 |
+
"num_tokens": 3394170.0,
|
| 578 |
+
"step": 480
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"entropy": 0.8164506537839771,
|
| 582 |
+
"epoch": 0.196,
|
| 583 |
+
"grad_norm": 54.25,
|
| 584 |
+
"learning_rate": 9.780000000000001e-06,
|
| 585 |
+
"loss": 13.1965,
|
| 586 |
+
"mean_token_accuracy": 0.7878578577190638,
|
| 587 |
+
"num_tokens": 3465341.0,
|
| 588 |
+
"step": 490
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"entropy": 0.8838530456647277,
|
| 592 |
+
"epoch": 0.2,
|
| 593 |
+
"grad_norm": 51.0,
|
| 594 |
+
"learning_rate": 9.980000000000001e-06,
|
| 595 |
+
"loss": 13.9771,
|
| 596 |
+
"mean_token_accuracy": 0.7734297584742308,
|
| 597 |
+
"num_tokens": 3533571.0,
|
| 598 |
+
"step": 500
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.2,
|
| 602 |
+
"eval_biology_entropy": 1.2344867270588875,
|
| 603 |
+
"eval_biology_loss": 1.2817195653915405,
|
| 604 |
+
"eval_biology_mean_token_accuracy": 0.6832386125922203,
|
| 605 |
+
"eval_biology_num_tokens": 3533571.0,
|
| 606 |
+
"eval_biology_runtime": 40.4092,
|
| 607 |
+
"eval_biology_samples_per_second": 12.373,
|
| 608 |
+
"eval_biology_steps_per_second": 12.373,
|
| 609 |
+
"step": 500
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.2,
|
| 613 |
+
"eval_math_entropy": 0.8546719763875008,
|
| 614 |
+
"eval_math_loss": 0.8547734618186951,
|
| 615 |
+
"eval_math_mean_token_accuracy": 0.7804626071453095,
|
| 616 |
+
"eval_math_num_tokens": 3533571.0,
|
| 617 |
+
"eval_math_runtime": 48.0961,
|
| 618 |
+
"eval_math_samples_per_second": 10.396,
|
| 619 |
+
"eval_math_steps_per_second": 10.396,
|
| 620 |
+
"step": 500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"entropy": 0.8578155474737287,
|
| 624 |
+
"epoch": 0.204,
|
| 625 |
+
"grad_norm": 45.75,
|
| 626 |
+
"learning_rate": 1.018e-05,
|
| 627 |
+
"loss": 13.8097,
|
| 628 |
+
"mean_token_accuracy": 0.776990968361497,
|
| 629 |
+
"num_tokens": 3602983.0,
|
| 630 |
+
"step": 510
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"entropy": 0.8306718476116657,
|
| 634 |
+
"epoch": 0.208,
|
| 635 |
+
"grad_norm": 50.25,
|
| 636 |
+
"learning_rate": 1.038e-05,
|
| 637 |
+
"loss": 13.4186,
|
| 638 |
+
"mean_token_accuracy": 0.7863588180392981,
|
| 639 |
+
"num_tokens": 3673600.0,
|
| 640 |
+
"step": 520
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"entropy": 0.8772016573697329,
|
| 644 |
+
"epoch": 0.212,
|
| 645 |
+
"grad_norm": 45.75,
|
| 646 |
+
"learning_rate": 1.0580000000000002e-05,
|
| 647 |
+
"loss": 13.9353,
|
| 648 |
+
"mean_token_accuracy": 0.7783012766391039,
|
| 649 |
+
"num_tokens": 3743641.0,
|
| 650 |
+
"step": 530
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"entropy": 0.8009312467649579,
|
| 654 |
+
"epoch": 0.216,
|
| 655 |
+
"grad_norm": 55.25,
|
| 656 |
+
"learning_rate": 1.0780000000000002e-05,
|
| 657 |
+
"loss": 12.9356,
|
| 658 |
+
"mean_token_accuracy": 0.7893661856651306,
|
| 659 |
+
"num_tokens": 3817397.0,
|
| 660 |
+
"step": 540
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"entropy": 0.8427618410438299,
|
| 664 |
+
"epoch": 0.22,
|
| 665 |
+
"grad_norm": 51.0,
|
| 666 |
+
"learning_rate": 1.0980000000000002e-05,
|
| 667 |
+
"loss": 13.3831,
|
| 668 |
+
"mean_token_accuracy": 0.7832117900252342,
|
| 669 |
+
"num_tokens": 3886866.0,
|
| 670 |
+
"step": 550
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"entropy": 0.8758323887363076,
|
| 674 |
+
"epoch": 0.224,
|
| 675 |
+
"grad_norm": 41.5,
|
| 676 |
+
"learning_rate": 1.1180000000000001e-05,
|
| 677 |
+
"loss": 13.8448,
|
| 678 |
+
"mean_token_accuracy": 0.7745687130838632,
|
| 679 |
+
"num_tokens": 3953732.0,
|
| 680 |
+
"step": 560
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"entropy": 0.8352954359725118,
|
| 684 |
+
"epoch": 0.228,
|
| 685 |
+
"grad_norm": 41.75,
|
| 686 |
+
"learning_rate": 1.138e-05,
|
| 687 |
+
"loss": 13.4861,
|
| 688 |
+
"mean_token_accuracy": 0.7828061260282994,
|
| 689 |
+
"num_tokens": 4025795.0,
|
| 690 |
+
"step": 570
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"entropy": 0.7947553901001811,
|
| 694 |
+
"epoch": 0.232,
|
| 695 |
+
"grad_norm": 43.25,
|
| 696 |
+
"learning_rate": 1.1580000000000001e-05,
|
| 697 |
+
"loss": 12.8366,
|
| 698 |
+
"mean_token_accuracy": 0.7880385994911194,
|
| 699 |
+
"num_tokens": 4098989.0,
|
| 700 |
+
"step": 580
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"entropy": 0.8285046070814133,
|
| 704 |
+
"epoch": 0.236,
|
| 705 |
+
"grad_norm": 39.25,
|
| 706 |
+
"learning_rate": 1.178e-05,
|
| 707 |
+
"loss": 13.0789,
|
| 708 |
+
"mean_token_accuracy": 0.7872951343655586,
|
| 709 |
+
"num_tokens": 4170878.0,
|
| 710 |
+
"step": 590
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"entropy": 0.7863880429416895,
|
| 714 |
+
"epoch": 0.24,
|
| 715 |
+
"grad_norm": 45.0,
|
| 716 |
+
"learning_rate": 1.198e-05,
|
| 717 |
+
"loss": 12.7479,
|
| 718 |
+
"mean_token_accuracy": 0.7928395148366689,
|
| 719 |
+
"num_tokens": 4243655.0,
|
| 720 |
+
"step": 600
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.24,
|
| 724 |
+
"eval_biology_entropy": 1.2514844969511032,
|
| 725 |
+
"eval_biology_loss": 1.2855682373046875,
|
| 726 |
+
"eval_biology_mean_token_accuracy": 0.6817903207540512,
|
| 727 |
+
"eval_biology_num_tokens": 4243655.0,
|
| 728 |
+
"eval_biology_runtime": 40.7639,
|
| 729 |
+
"eval_biology_samples_per_second": 12.266,
|
| 730 |
+
"eval_biology_steps_per_second": 12.266,
|
| 731 |
+
"step": 600
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.24,
|
| 735 |
+
"eval_math_entropy": 0.843876126229763,
|
| 736 |
+
"eval_math_loss": 0.8332411646842957,
|
| 737 |
+
"eval_math_mean_token_accuracy": 0.783966186761856,
|
| 738 |
+
"eval_math_num_tokens": 4243655.0,
|
| 739 |
+
"eval_math_runtime": 48.1965,
|
| 740 |
+
"eval_math_samples_per_second": 10.374,
|
| 741 |
+
"eval_math_steps_per_second": 10.374,
|
| 742 |
+
"step": 600
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"entropy": 0.7838549023494125,
|
| 746 |
+
"epoch": 0.244,
|
| 747 |
+
"grad_norm": 41.75,
|
| 748 |
+
"learning_rate": 1.218e-05,
|
| 749 |
+
"loss": 12.4042,
|
| 750 |
+
"mean_token_accuracy": 0.7971927735954523,
|
| 751 |
+
"num_tokens": 4317321.0,
|
| 752 |
+
"step": 610
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"entropy": 0.7797296568751335,
|
| 756 |
+
"epoch": 0.248,
|
| 757 |
+
"grad_norm": 51.75,
|
| 758 |
+
"learning_rate": 1.2380000000000002e-05,
|
| 759 |
+
"loss": 12.5895,
|
| 760 |
+
"mean_token_accuracy": 0.7916548330336809,
|
| 761 |
+
"num_tokens": 4387651.0,
|
| 762 |
+
"step": 620
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"entropy": 0.8094759196043014,
|
| 766 |
+
"epoch": 0.252,
|
| 767 |
+
"grad_norm": 84.5,
|
| 768 |
+
"learning_rate": 1.2580000000000002e-05,
|
| 769 |
+
"loss": 12.9446,
|
| 770 |
+
"mean_token_accuracy": 0.7892590902745724,
|
| 771 |
+
"num_tokens": 4461909.0,
|
| 772 |
+
"step": 630
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"entropy": 0.7797345662489533,
|
| 776 |
+
"epoch": 0.256,
|
| 777 |
+
"grad_norm": 62.0,
|
| 778 |
+
"learning_rate": 1.2780000000000001e-05,
|
| 779 |
+
"loss": 12.462,
|
| 780 |
+
"mean_token_accuracy": 0.7935512393712998,
|
| 781 |
+
"num_tokens": 4531471.0,
|
| 782 |
+
"step": 640
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"entropy": 0.7911159221082926,
|
| 786 |
+
"epoch": 0.26,
|
| 787 |
+
"grad_norm": 40.0,
|
| 788 |
+
"learning_rate": 1.2980000000000001e-05,
|
| 789 |
+
"loss": 12.8515,
|
| 790 |
+
"mean_token_accuracy": 0.7928617469966411,
|
| 791 |
+
"num_tokens": 4600757.0,
|
| 792 |
+
"step": 650
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"entropy": 0.8169831104576588,
|
| 796 |
+
"epoch": 0.264,
|
| 797 |
+
"grad_norm": 49.5,
|
| 798 |
+
"learning_rate": 1.3180000000000001e-05,
|
| 799 |
+
"loss": 13.117,
|
| 800 |
+
"mean_token_accuracy": 0.7893466159701348,
|
| 801 |
+
"num_tokens": 4669389.0,
|
| 802 |
+
"step": 660
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"entropy": 0.8287884982302784,
|
| 806 |
+
"epoch": 0.268,
|
| 807 |
+
"grad_norm": 55.25,
|
| 808 |
+
"learning_rate": 1.3380000000000002e-05,
|
| 809 |
+
"loss": 13.3618,
|
| 810 |
+
"mean_token_accuracy": 0.7853043630719185,
|
| 811 |
+
"num_tokens": 4740492.0,
|
| 812 |
+
"step": 670
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"entropy": 0.7938818972557783,
|
| 816 |
+
"epoch": 0.272,
|
| 817 |
+
"grad_norm": 42.5,
|
| 818 |
+
"learning_rate": 1.3580000000000002e-05,
|
| 819 |
+
"loss": 12.6092,
|
| 820 |
+
"mean_token_accuracy": 0.7941738195717335,
|
| 821 |
+
"num_tokens": 4810284.0,
|
| 822 |
+
"step": 680
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"entropy": 0.7964617364108563,
|
| 826 |
+
"epoch": 0.276,
|
| 827 |
+
"grad_norm": 37.25,
|
| 828 |
+
"learning_rate": 1.378e-05,
|
| 829 |
+
"loss": 12.6495,
|
| 830 |
+
"mean_token_accuracy": 0.7921258114278317,
|
| 831 |
+
"num_tokens": 4885701.0,
|
| 832 |
+
"step": 690
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"entropy": 0.7932526333257556,
|
| 836 |
+
"epoch": 0.28,
|
| 837 |
+
"grad_norm": 48.5,
|
| 838 |
+
"learning_rate": 1.398e-05,
|
| 839 |
+
"loss": 12.8488,
|
| 840 |
+
"mean_token_accuracy": 0.7890512771904469,
|
| 841 |
+
"num_tokens": 4955375.0,
|
| 842 |
+
"step": 700
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.28,
|
| 846 |
+
"eval_biology_entropy": 1.2677810794115068,
|
| 847 |
+
"eval_biology_loss": 1.2936662435531616,
|
| 848 |
+
"eval_biology_mean_token_accuracy": 0.6814599596261978,
|
| 849 |
+
"eval_biology_num_tokens": 4955375.0,
|
| 850 |
+
"eval_biology_runtime": 40.9028,
|
| 851 |
+
"eval_biology_samples_per_second": 12.224,
|
| 852 |
+
"eval_biology_steps_per_second": 12.224,
|
| 853 |
+
"step": 700
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 0.28,
|
| 857 |
+
"eval_math_entropy": 0.8274374225139618,
|
| 858 |
+
"eval_math_loss": 0.8181362748146057,
|
| 859 |
+
"eval_math_mean_token_accuracy": 0.7868153600692749,
|
| 860 |
+
"eval_math_num_tokens": 4955375.0,
|
| 861 |
+
"eval_math_runtime": 48.2074,
|
| 862 |
+
"eval_math_samples_per_second": 10.372,
|
| 863 |
+
"eval_math_steps_per_second": 10.372,
|
| 864 |
+
"step": 700
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"entropy": 0.7877025598660111,
|
| 868 |
+
"epoch": 0.284,
|
| 869 |
+
"grad_norm": 39.25,
|
| 870 |
+
"learning_rate": 1.418e-05,
|
| 871 |
+
"loss": 12.6293,
|
| 872 |
+
"mean_token_accuracy": 0.7944098394364119,
|
| 873 |
+
"num_tokens": 5025638.0,
|
| 874 |
+
"step": 710
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"entropy": 0.7564179511740804,
|
| 878 |
+
"epoch": 0.288,
|
| 879 |
+
"grad_norm": 38.75,
|
| 880 |
+
"learning_rate": 1.4380000000000001e-05,
|
| 881 |
+
"loss": 12.0294,
|
| 882 |
+
"mean_token_accuracy": 0.8003660906106234,
|
| 883 |
+
"num_tokens": 5095104.0,
|
| 884 |
+
"step": 720
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"entropy": 0.7647898152470589,
|
| 888 |
+
"epoch": 0.292,
|
| 889 |
+
"grad_norm": 56.0,
|
| 890 |
+
"learning_rate": 1.4580000000000001e-05,
|
| 891 |
+
"loss": 12.4713,
|
| 892 |
+
"mean_token_accuracy": 0.7947716873139143,
|
| 893 |
+
"num_tokens": 5167049.0,
|
| 894 |
+
"step": 730
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"entropy": 0.8119000235572458,
|
| 898 |
+
"epoch": 0.296,
|
| 899 |
+
"grad_norm": 36.75,
|
| 900 |
+
"learning_rate": 1.478e-05,
|
| 901 |
+
"loss": 13.0034,
|
| 902 |
+
"mean_token_accuracy": 0.7876484204083681,
|
| 903 |
+
"num_tokens": 5240988.0,
|
| 904 |
+
"step": 740
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"entropy": 0.7802535065449774,
|
| 908 |
+
"epoch": 0.3,
|
| 909 |
+
"grad_norm": 40.5,
|
| 910 |
+
"learning_rate": 1.498e-05,
|
| 911 |
+
"loss": 12.4371,
|
| 912 |
+
"mean_token_accuracy": 0.795463678240776,
|
| 913 |
+
"num_tokens": 5312258.0,
|
| 914 |
+
"step": 750
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"entropy": 0.8087782407179475,
|
| 918 |
+
"epoch": 0.304,
|
| 919 |
+
"grad_norm": 44.75,
|
| 920 |
+
"learning_rate": 1.5180000000000002e-05,
|
| 921 |
+
"loss": 12.9661,
|
| 922 |
+
"mean_token_accuracy": 0.7857579983770847,
|
| 923 |
+
"num_tokens": 5383732.0,
|
| 924 |
+
"step": 760
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"entropy": 0.8243673056364059,
|
| 928 |
+
"epoch": 0.308,
|
| 929 |
+
"grad_norm": 39.5,
|
| 930 |
+
"learning_rate": 1.5380000000000002e-05,
|
| 931 |
+
"loss": 13.1334,
|
| 932 |
+
"mean_token_accuracy": 0.7867084331810474,
|
| 933 |
+
"num_tokens": 5456650.0,
|
| 934 |
+
"step": 770
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"entropy": 0.7901528097689152,
|
| 938 |
+
"epoch": 0.312,
|
| 939 |
+
"grad_norm": 40.0,
|
| 940 |
+
"learning_rate": 1.5580000000000003e-05,
|
| 941 |
+
"loss": 12.6608,
|
| 942 |
+
"mean_token_accuracy": 0.7905277730897069,
|
| 943 |
+
"num_tokens": 5530666.0,
|
| 944 |
+
"step": 780
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"entropy": 0.7515985878184438,
|
| 948 |
+
"epoch": 0.316,
|
| 949 |
+
"grad_norm": 44.75,
|
| 950 |
+
"learning_rate": 1.578e-05,
|
| 951 |
+
"loss": 11.9655,
|
| 952 |
+
"mean_token_accuracy": 0.8011779848486185,
|
| 953 |
+
"num_tokens": 5605269.0,
|
| 954 |
+
"step": 790
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"entropy": 0.7314818985760212,
|
| 958 |
+
"epoch": 0.32,
|
| 959 |
+
"grad_norm": 42.5,
|
| 960 |
+
"learning_rate": 1.5980000000000003e-05,
|
| 961 |
+
"loss": 11.7244,
|
| 962 |
+
"mean_token_accuracy": 0.8059452049434185,
|
| 963 |
+
"num_tokens": 5676334.0,
|
| 964 |
+
"step": 800
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 0.32,
|
| 968 |
+
"eval_biology_entropy": 1.2663928886651994,
|
| 969 |
+
"eval_biology_loss": 1.2994052171707153,
|
| 970 |
+
"eval_biology_mean_token_accuracy": 0.6803199350833893,
|
| 971 |
+
"eval_biology_num_tokens": 5676334.0,
|
| 972 |
+
"eval_biology_runtime": 40.7632,
|
| 973 |
+
"eval_biology_samples_per_second": 12.266,
|
| 974 |
+
"eval_biology_steps_per_second": 12.266,
|
| 975 |
+
"step": 800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.32,
|
| 979 |
+
"eval_math_entropy": 0.7805578476190567,
|
| 980 |
+
"eval_math_loss": 0.8004927635192871,
|
| 981 |
+
"eval_math_mean_token_accuracy": 0.7903167794942856,
|
| 982 |
+
"eval_math_num_tokens": 5676334.0,
|
| 983 |
+
"eval_math_runtime": 48.1783,
|
| 984 |
+
"eval_math_samples_per_second": 10.378,
|
| 985 |
+
"eval_math_steps_per_second": 10.378,
|
| 986 |
+
"step": 800
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"entropy": 0.7661854576319456,
|
| 990 |
+
"epoch": 0.324,
|
| 991 |
+
"grad_norm": 40.5,
|
| 992 |
+
"learning_rate": 1.618e-05,
|
| 993 |
+
"loss": 12.5767,
|
| 994 |
+
"mean_token_accuracy": 0.7944763712584972,
|
| 995 |
+
"num_tokens": 5747779.0,
|
| 996 |
+
"step": 810
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"entropy": 0.803387301787734,
|
| 1000 |
+
"epoch": 0.328,
|
| 1001 |
+
"grad_norm": 42.0,
|
| 1002 |
+
"learning_rate": 1.638e-05,
|
| 1003 |
+
"loss": 12.825,
|
| 1004 |
+
"mean_token_accuracy": 0.7907922245562077,
|
| 1005 |
+
"num_tokens": 5816488.0,
|
| 1006 |
+
"step": 820
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"entropy": 0.8049221463501454,
|
| 1010 |
+
"epoch": 0.332,
|
| 1011 |
+
"grad_norm": 44.0,
|
| 1012 |
+
"learning_rate": 1.658e-05,
|
| 1013 |
+
"loss": 12.9669,
|
| 1014 |
+
"mean_token_accuracy": 0.7854703675955534,
|
| 1015 |
+
"num_tokens": 5887180.0,
|
| 1016 |
+
"step": 830
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"entropy": 0.8102014064788818,
|
| 1020 |
+
"epoch": 0.336,
|
| 1021 |
+
"grad_norm": 38.5,
|
| 1022 |
+
"learning_rate": 1.6780000000000002e-05,
|
| 1023 |
+
"loss": 12.9814,
|
| 1024 |
+
"mean_token_accuracy": 0.7886964596807957,
|
| 1025 |
+
"num_tokens": 5958480.0,
|
| 1026 |
+
"step": 840
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"entropy": 0.7344874950125814,
|
| 1030 |
+
"epoch": 0.34,
|
| 1031 |
+
"grad_norm": 50.25,
|
| 1032 |
+
"learning_rate": 1.698e-05,
|
| 1033 |
+
"loss": 11.7609,
|
| 1034 |
+
"mean_token_accuracy": 0.8062219381332397,
|
| 1035 |
+
"num_tokens": 6031020.0,
|
| 1036 |
+
"step": 850
|
| 1037 |
+
},
|
| 1038 |
+
{
|
| 1039 |
+
"entropy": 0.802255941182375,
|
| 1040 |
+
"epoch": 0.344,
|
| 1041 |
+
"grad_norm": 39.0,
|
| 1042 |
+
"learning_rate": 1.718e-05,
|
| 1043 |
+
"loss": 12.8439,
|
| 1044 |
+
"mean_token_accuracy": 0.7877463910728693,
|
| 1045 |
+
"num_tokens": 6102081.0,
|
| 1046 |
+
"step": 860
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"entropy": 0.7980229431763292,
|
| 1050 |
+
"epoch": 0.348,
|
| 1051 |
+
"grad_norm": 36.25,
|
| 1052 |
+
"learning_rate": 1.7380000000000003e-05,
|
| 1053 |
+
"loss": 12.7774,
|
| 1054 |
+
"mean_token_accuracy": 0.7900870595127344,
|
| 1055 |
+
"num_tokens": 6172871.0,
|
| 1056 |
+
"step": 870
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"entropy": 0.7401213280856609,
|
| 1060 |
+
"epoch": 0.352,
|
| 1061 |
+
"grad_norm": 40.75,
|
| 1062 |
+
"learning_rate": 1.758e-05,
|
| 1063 |
+
"loss": 11.9409,
|
| 1064 |
+
"mean_token_accuracy": 0.801040057092905,
|
| 1065 |
+
"num_tokens": 6242161.0,
|
| 1066 |
+
"step": 880
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"entropy": 0.8095649138092995,
|
| 1070 |
+
"epoch": 0.356,
|
| 1071 |
+
"grad_norm": 46.5,
|
| 1072 |
+
"learning_rate": 1.7780000000000003e-05,
|
| 1073 |
+
"loss": 12.9134,
|
| 1074 |
+
"mean_token_accuracy": 0.7872339155524969,
|
| 1075 |
+
"num_tokens": 6312187.0,
|
| 1076 |
+
"step": 890
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"entropy": 0.7853369496762752,
|
| 1080 |
+
"epoch": 0.36,
|
| 1081 |
+
"grad_norm": 37.75,
|
| 1082 |
+
"learning_rate": 1.798e-05,
|
| 1083 |
+
"loss": 12.4062,
|
| 1084 |
+
"mean_token_accuracy": 0.7960268270224333,
|
| 1085 |
+
"num_tokens": 6377327.0,
|
| 1086 |
+
"step": 900
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.36,
|
| 1090 |
+
"eval_biology_entropy": 1.250610340833664,
|
| 1091 |
+
"eval_biology_loss": 1.3047173023223877,
|
| 1092 |
+
"eval_biology_mean_token_accuracy": 0.6794998321533203,
|
| 1093 |
+
"eval_biology_num_tokens": 6377327.0,
|
| 1094 |
+
"eval_biology_runtime": 40.5105,
|
| 1095 |
+
"eval_biology_samples_per_second": 12.342,
|
| 1096 |
+
"eval_biology_steps_per_second": 12.342,
|
| 1097 |
+
"step": 900
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.36,
|
| 1101 |
+
"eval_math_entropy": 0.773322255730629,
|
| 1102 |
+
"eval_math_loss": 0.789590060710907,
|
| 1103 |
+
"eval_math_mean_token_accuracy": 0.7914780566692352,
|
| 1104 |
+
"eval_math_num_tokens": 6377327.0,
|
| 1105 |
+
"eval_math_runtime": 48.04,
|
| 1106 |
+
"eval_math_samples_per_second": 10.408,
|
| 1107 |
+
"eval_math_steps_per_second": 10.408,
|
| 1108 |
+
"step": 900
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"entropy": 0.763472201116383,
|
| 1112 |
+
"epoch": 0.364,
|
| 1113 |
+
"grad_norm": 41.5,
|
| 1114 |
+
"learning_rate": 1.8180000000000002e-05,
|
| 1115 |
+
"loss": 12.3288,
|
| 1116 |
+
"mean_token_accuracy": 0.7959023587405681,
|
| 1117 |
+
"num_tokens": 6453254.0,
|
| 1118 |
+
"step": 910
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"entropy": 0.7712945869192481,
|
| 1122 |
+
"epoch": 0.368,
|
| 1123 |
+
"grad_norm": 40.25,
|
| 1124 |
+
"learning_rate": 1.8380000000000004e-05,
|
| 1125 |
+
"loss": 12.2854,
|
| 1126 |
+
"mean_token_accuracy": 0.795868669450283,
|
| 1127 |
+
"num_tokens": 6523679.0,
|
| 1128 |
+
"step": 920
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"entropy": 0.7641221083700657,
|
| 1132 |
+
"epoch": 0.372,
|
| 1133 |
+
"grad_norm": 37.25,
|
| 1134 |
+
"learning_rate": 1.858e-05,
|
| 1135 |
+
"loss": 12.2867,
|
| 1136 |
+
"mean_token_accuracy": 0.7949323236942292,
|
| 1137 |
+
"num_tokens": 6594651.0,
|
| 1138 |
+
"step": 930
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"entropy": 0.7518373979255557,
|
| 1142 |
+
"epoch": 0.376,
|
| 1143 |
+
"grad_norm": 42.0,
|
| 1144 |
+
"learning_rate": 1.878e-05,
|
| 1145 |
+
"loss": 12.1179,
|
| 1146 |
+
"mean_token_accuracy": 0.798585458099842,
|
| 1147 |
+
"num_tokens": 6669278.0,
|
| 1148 |
+
"step": 940
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"entropy": 0.7543134400621057,
|
| 1152 |
+
"epoch": 0.38,
|
| 1153 |
+
"grad_norm": 33.75,
|
| 1154 |
+
"learning_rate": 1.898e-05,
|
| 1155 |
+
"loss": 12.023,
|
| 1156 |
+
"mean_token_accuracy": 0.7991646058857441,
|
| 1157 |
+
"num_tokens": 6740491.0,
|
| 1158 |
+
"step": 950
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"entropy": 0.7203706834465265,
|
| 1162 |
+
"epoch": 0.384,
|
| 1163 |
+
"grad_norm": 35.5,
|
| 1164 |
+
"learning_rate": 1.918e-05,
|
| 1165 |
+
"loss": 11.6634,
|
| 1166 |
+
"mean_token_accuracy": 0.8036353968083858,
|
| 1167 |
+
"num_tokens": 6810978.0,
|
| 1168 |
+
"step": 960
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"entropy": 0.7761434537358582,
|
| 1172 |
+
"epoch": 0.388,
|
| 1173 |
+
"grad_norm": 39.5,
|
| 1174 |
+
"learning_rate": 1.938e-05,
|
| 1175 |
+
"loss": 12.4387,
|
| 1176 |
+
"mean_token_accuracy": 0.7944561909884215,
|
| 1177 |
+
"num_tokens": 6882077.0,
|
| 1178 |
+
"step": 970
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"entropy": 0.7785217678174376,
|
| 1182 |
+
"epoch": 0.392,
|
| 1183 |
+
"grad_norm": 33.75,
|
| 1184 |
+
"learning_rate": 1.9580000000000002e-05,
|
| 1185 |
+
"loss": 12.607,
|
| 1186 |
+
"mean_token_accuracy": 0.7944069147109986,
|
| 1187 |
+
"num_tokens": 6953929.0,
|
| 1188 |
+
"step": 980
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"entropy": 0.8195465696975589,
|
| 1192 |
+
"epoch": 0.396,
|
| 1193 |
+
"grad_norm": 65.5,
|
| 1194 |
+
"learning_rate": 1.978e-05,
|
| 1195 |
+
"loss": 13.2202,
|
| 1196 |
+
"mean_token_accuracy": 0.7838506288826466,
|
| 1197 |
+
"num_tokens": 7023835.0,
|
| 1198 |
+
"step": 990
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"entropy": 0.8217529645189643,
|
| 1202 |
+
"epoch": 0.4,
|
| 1203 |
+
"grad_norm": 39.75,
|
| 1204 |
+
"learning_rate": 1.9980000000000002e-05,
|
| 1205 |
+
"loss": 13.1698,
|
| 1206 |
+
"mean_token_accuracy": 0.7848756689578295,
|
| 1207 |
+
"num_tokens": 7096903.0,
|
| 1208 |
+
"step": 1000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.4,
|
| 1212 |
+
"eval_biology_entropy": 1.266352553486824,
|
| 1213 |
+
"eval_biology_loss": 1.312301516532898,
|
| 1214 |
+
"eval_biology_mean_token_accuracy": 0.6775026069879532,
|
| 1215 |
+
"eval_biology_num_tokens": 7096903.0,
|
| 1216 |
+
"eval_biology_runtime": 40.7343,
|
| 1217 |
+
"eval_biology_samples_per_second": 12.275,
|
| 1218 |
+
"eval_biology_steps_per_second": 12.275,
|
| 1219 |
+
"step": 1000
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.4,
|
| 1223 |
+
"eval_math_entropy": 0.7719556384682655,
|
| 1224 |
+
"eval_math_loss": 0.7821868062019348,
|
| 1225 |
+
"eval_math_mean_token_accuracy": 0.7932658619880676,
|
| 1226 |
+
"eval_math_num_tokens": 7096903.0,
|
| 1227 |
+
"eval_math_runtime": 47.8946,
|
| 1228 |
+
"eval_math_samples_per_second": 10.44,
|
| 1229 |
+
"eval_math_steps_per_second": 10.44,
|
| 1230 |
+
"step": 1000
|
| 1231 |
+
}
|
| 1232 |
+
],
|
| 1233 |
+
"logging_steps": 10,
|
| 1234 |
+
"max_steps": 10000,
|
| 1235 |
+
"num_input_tokens_seen": 0,
|
| 1236 |
+
"num_train_epochs": 4,
|
| 1237 |
+
"save_steps": 500,
|
| 1238 |
+
"stateful_callbacks": {
|
| 1239 |
+
"TrainerControl": {
|
| 1240 |
+
"args": {
|
| 1241 |
+
"should_epoch_stop": false,
|
| 1242 |
+
"should_evaluate": false,
|
| 1243 |
+
"should_log": false,
|
| 1244 |
+
"should_save": true,
|
| 1245 |
+
"should_training_stop": false
|
| 1246 |
+
},
|
| 1247 |
+
"attributes": {}
|
| 1248 |
+
}
|
| 1249 |
+
},
|
| 1250 |
+
"total_flos": 4.7587833384015814e+17,
|
| 1251 |
+
"train_batch_size": 1,
|
| 1252 |
+
"trial_name": null,
|
| 1253 |
+
"trial_params": null
|
| 1254 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6df9046062202b486a4fcb0ad2a415dd6b0a1a521e0667c7f40799812ff8d06e
|
| 3 |
+
size 6481
|