Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- checkpoint-1500/added_tokens.json +3 -0
- checkpoint-1500/chat_template.jinja +47 -0
- checkpoint-1500/config.json +110 -0
- checkpoint-1500/generation_config.json +13 -0
- checkpoint-1500/model-00001-of-00005.safetensors +3 -0
- checkpoint-1500/model-00002-of-00005.safetensors +3 -0
- checkpoint-1500/model-00003-of-00005.safetensors +3 -0
- checkpoint-1500/model-00004-of-00005.safetensors +3 -0
- checkpoint-1500/model-00005-of-00005.safetensors +3 -0
- checkpoint-1500/model.safetensors.index.json +0 -0
- checkpoint-1500/optimizer.pt +3 -0
- checkpoint-1500/rng_state.pth +3 -0
- checkpoint-1500/scheduler.pt +3 -0
- checkpoint-1500/special_tokens_map.json +33 -0
- checkpoint-1500/tokenizer.json +3 -0
- checkpoint-1500/tokenizer.model +3 -0
- checkpoint-1500/tokenizer_config.json +0 -0
- checkpoint-1500/trainer_state.json +2194 -0
- checkpoint-1500/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-1500/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image_soft_token>": 262144
|
| 3 |
+
}
|
checkpoint-1500/chat_template.jinja
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{{ bos_token }}
|
| 2 |
+
{%- if messages[0]['role'] == 'system' -%}
|
| 3 |
+
{%- if messages[0]['content'] is string -%}
|
| 4 |
+
{%- set first_user_prefix = messages[0]['content'] + '
|
| 5 |
+
|
| 6 |
+
' -%}
|
| 7 |
+
{%- else -%}
|
| 8 |
+
{%- set first_user_prefix = messages[0]['content'][0]['text'] + '
|
| 9 |
+
|
| 10 |
+
' -%}
|
| 11 |
+
{%- endif -%}
|
| 12 |
+
{%- set loop_messages = messages[1:] -%}
|
| 13 |
+
{%- else -%}
|
| 14 |
+
{%- set first_user_prefix = "" -%}
|
| 15 |
+
{%- set loop_messages = messages -%}
|
| 16 |
+
{%- endif -%}
|
| 17 |
+
{%- for message in loop_messages -%}
|
| 18 |
+
{%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
|
| 19 |
+
{{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
|
| 20 |
+
{%- endif -%}
|
| 21 |
+
{%- if (message['role'] == 'assistant') -%}
|
| 22 |
+
{%- set role = "model" -%}
|
| 23 |
+
{%- else -%}
|
| 24 |
+
{%- set role = message['role'] -%}
|
| 25 |
+
{%- endif -%}
|
| 26 |
+
{{ '<start_of_turn>' + role + '
|
| 27 |
+
' + (first_user_prefix if loop.first else "") }}
|
| 28 |
+
{%- if message['content'] is string -%}
|
| 29 |
+
{{ message['content'] | trim }}
|
| 30 |
+
{%- elif message['content'] is iterable -%}
|
| 31 |
+
{%- for item in message['content'] -%}
|
| 32 |
+
{%- if item['type'] == 'image' -%}
|
| 33 |
+
{{ '<start_of_image>' }}
|
| 34 |
+
{%- elif item['type'] == 'text' -%}
|
| 35 |
+
{{ item['text'] | trim }}
|
| 36 |
+
{%- endif -%}
|
| 37 |
+
{%- endfor -%}
|
| 38 |
+
{%- else -%}
|
| 39 |
+
{{ raise_exception("Invalid content type") }}
|
| 40 |
+
{%- endif -%}
|
| 41 |
+
{{ '<end_of_turn>
|
| 42 |
+
' }}
|
| 43 |
+
{%- endfor -%}
|
| 44 |
+
{%- if add_generation_prompt -%}
|
| 45 |
+
{{'<start_of_turn>model
|
| 46 |
+
'}}
|
| 47 |
+
{%- endif -%}
|
checkpoint-1500/config.json
ADDED
|
@@ -0,0 +1,110 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Gemma3ForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"boi_token_index": 255999,
|
| 6 |
+
"bos_token_id": 2,
|
| 7 |
+
"dtype": "bfloat16",
|
| 8 |
+
"eoi_token_index": 256000,
|
| 9 |
+
"eos_token_id": 1,
|
| 10 |
+
"image_token_index": 262144,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"mm_tokens_per_image": 256,
|
| 13 |
+
"model_type": "gemma3",
|
| 14 |
+
"pad_token_id": 0,
|
| 15 |
+
"text_config": {
|
| 16 |
+
"_sliding_window_pattern": 6,
|
| 17 |
+
"attention_bias": false,
|
| 18 |
+
"attention_dropout": 0.0,
|
| 19 |
+
"attn_logit_softcapping": null,
|
| 20 |
+
"dtype": "bfloat16",
|
| 21 |
+
"final_logit_softcapping": null,
|
| 22 |
+
"head_dim": 256,
|
| 23 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
| 24 |
+
"hidden_size": 3840,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"intermediate_size": 15360,
|
| 27 |
+
"layer_types": [
|
| 28 |
+
"sliding_attention",
|
| 29 |
+
"sliding_attention",
|
| 30 |
+
"sliding_attention",
|
| 31 |
+
"sliding_attention",
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"full_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"sliding_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"sliding_attention",
|
| 38 |
+
"sliding_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"sliding_attention",
|
| 41 |
+
"sliding_attention",
|
| 42 |
+
"sliding_attention",
|
| 43 |
+
"sliding_attention",
|
| 44 |
+
"sliding_attention",
|
| 45 |
+
"full_attention",
|
| 46 |
+
"sliding_attention",
|
| 47 |
+
"sliding_attention",
|
| 48 |
+
"sliding_attention",
|
| 49 |
+
"sliding_attention",
|
| 50 |
+
"sliding_attention",
|
| 51 |
+
"full_attention",
|
| 52 |
+
"sliding_attention",
|
| 53 |
+
"sliding_attention",
|
| 54 |
+
"sliding_attention",
|
| 55 |
+
"sliding_attention",
|
| 56 |
+
"sliding_attention",
|
| 57 |
+
"full_attention",
|
| 58 |
+
"sliding_attention",
|
| 59 |
+
"sliding_attention",
|
| 60 |
+
"sliding_attention",
|
| 61 |
+
"sliding_attention",
|
| 62 |
+
"sliding_attention",
|
| 63 |
+
"full_attention",
|
| 64 |
+
"sliding_attention",
|
| 65 |
+
"sliding_attention",
|
| 66 |
+
"sliding_attention",
|
| 67 |
+
"sliding_attention",
|
| 68 |
+
"sliding_attention",
|
| 69 |
+
"full_attention",
|
| 70 |
+
"sliding_attention",
|
| 71 |
+
"sliding_attention",
|
| 72 |
+
"sliding_attention",
|
| 73 |
+
"sliding_attention",
|
| 74 |
+
"sliding_attention",
|
| 75 |
+
"full_attention"
|
| 76 |
+
],
|
| 77 |
+
"max_position_embeddings": 131072,
|
| 78 |
+
"model_type": "gemma3_text",
|
| 79 |
+
"num_attention_heads": 16,
|
| 80 |
+
"num_hidden_layers": 48,
|
| 81 |
+
"num_key_value_heads": 8,
|
| 82 |
+
"query_pre_attn_scalar": 256,
|
| 83 |
+
"rms_norm_eps": 1e-06,
|
| 84 |
+
"rope_local_base_freq": 10000.0,
|
| 85 |
+
"rope_scaling": {
|
| 86 |
+
"factor": 8.0,
|
| 87 |
+
"rope_type": "linear"
|
| 88 |
+
},
|
| 89 |
+
"rope_theta": 1000000.0,
|
| 90 |
+
"sliding_window": 1024,
|
| 91 |
+
"use_cache": true,
|
| 92 |
+
"vocab_size": 262208
|
| 93 |
+
},
|
| 94 |
+
"transformers_version": "4.56.1",
|
| 95 |
+
"vision_config": {
|
| 96 |
+
"attention_dropout": 0.0,
|
| 97 |
+
"dtype": "bfloat16",
|
| 98 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 99 |
+
"hidden_size": 1152,
|
| 100 |
+
"image_size": 896,
|
| 101 |
+
"intermediate_size": 4304,
|
| 102 |
+
"layer_norm_eps": 1e-06,
|
| 103 |
+
"model_type": "siglip_vision_model",
|
| 104 |
+
"num_attention_heads": 16,
|
| 105 |
+
"num_channels": 3,
|
| 106 |
+
"num_hidden_layers": 27,
|
| 107 |
+
"patch_size": 14,
|
| 108 |
+
"vision_use_head": false
|
| 109 |
+
}
|
| 110 |
+
}
|
checkpoint-1500/generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 2,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
1,
|
| 6 |
+
1,
|
| 7 |
+
106
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 0,
|
| 10 |
+
"top_k": 64,
|
| 11 |
+
"top_p": 0.95,
|
| 12 |
+
"transformers_version": "4.56.1"
|
| 13 |
+
}
|
checkpoint-1500/model-00001-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4847447e92599833e8dbaa3067cd201c3bb5c052efa91f11ba891e43234f7832
|
| 3 |
+
size 4979902192
|
checkpoint-1500/model-00002-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:891bd54eed03cba9ee1e705533a02a8217fcc29f356e4a1f53e5fd0d178883ad
|
| 3 |
+
size 4931296592
|
checkpoint-1500/model-00003-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
|
| 3 |
+
size 4931296656
|
checkpoint-1500/model-00004-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
|
| 3 |
+
size 4931296656
|
checkpoint-1500/model-00005-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c96bc8976fc7a9a665227776bbc747c602785dbc7b21e62b74af4981fee193f
|
| 3 |
+
size 4601000928
|
checkpoint-1500/model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:316522e4b809317277ffdbcc81da90c12f5cd777ce58deccfdc091e769fa2472
|
| 3 |
+
size 896607944
|
checkpoint-1500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e18baad484d226f8ed089a4849ca70f9922b4c7d45ed627c4dfd43cb9bc6002
|
| 3 |
+
size 14645
|
checkpoint-1500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c06b4f8f3b3f6004cf9db5f25ea010ea45540a89813f901db1e65629d2290fe
|
| 3 |
+
size 1465
|
checkpoint-1500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"boi_token": "<start_of_image>",
|
| 3 |
+
"bos_token": {
|
| 4 |
+
"content": "<bos>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
"eoi_token": "<end_of_image>",
|
| 11 |
+
"eos_token": {
|
| 12 |
+
"content": "<eos>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false
|
| 17 |
+
},
|
| 18 |
+
"image_token": "<image_soft_token>",
|
| 19 |
+
"pad_token": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false
|
| 25 |
+
},
|
| 26 |
+
"unk_token": {
|
| 27 |
+
"content": "<unk>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": false,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
}
|
| 33 |
+
}
|
checkpoint-1500/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
|
| 3 |
+
size 33384568
|
checkpoint-1500/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
|
| 3 |
+
size 4689074
|
checkpoint-1500/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/trainer_state.json
ADDED
|
@@ -0,0 +1,2194 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 2.4,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 0.7213201837614178,
|
| 14 |
+
"epoch": 0.016,
|
| 15 |
+
"grad_norm": 127.0,
|
| 16 |
+
"learning_rate": 6.000000000000001e-07,
|
| 17 |
+
"loss": 33.4275,
|
| 18 |
+
"mean_token_accuracy": 0.6860143840312958,
|
| 19 |
+
"num_tokens": 280941.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 0.75641950070858,
|
| 24 |
+
"epoch": 0.032,
|
| 25 |
+
"grad_norm": 118.0,
|
| 26 |
+
"learning_rate": 1.2666666666666669e-06,
|
| 27 |
+
"loss": 33.8085,
|
| 28 |
+
"mean_token_accuracy": 0.6810433834791183,
|
| 29 |
+
"num_tokens": 558056.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 0.7386618180200457,
|
| 34 |
+
"epoch": 0.048,
|
| 35 |
+
"grad_norm": 105.5,
|
| 36 |
+
"learning_rate": 1.9333333333333336e-06,
|
| 37 |
+
"loss": 33.0823,
|
| 38 |
+
"mean_token_accuracy": 0.686032148450613,
|
| 39 |
+
"num_tokens": 836753.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 0.7653292674571276,
|
| 44 |
+
"epoch": 0.064,
|
| 45 |
+
"grad_norm": 106.0,
|
| 46 |
+
"learning_rate": 2.6e-06,
|
| 47 |
+
"loss": 33.3421,
|
| 48 |
+
"mean_token_accuracy": 0.678343191742897,
|
| 49 |
+
"num_tokens": 1126446.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 0.7746079994365573,
|
| 54 |
+
"epoch": 0.08,
|
| 55 |
+
"grad_norm": 112.0,
|
| 56 |
+
"learning_rate": 3.266666666666667e-06,
|
| 57 |
+
"loss": 31.8715,
|
| 58 |
+
"mean_token_accuracy": 0.6860048722475767,
|
| 59 |
+
"num_tokens": 1413596.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 0.7967683909460902,
|
| 64 |
+
"epoch": 0.096,
|
| 65 |
+
"grad_norm": 87.5,
|
| 66 |
+
"learning_rate": 3.9333333333333335e-06,
|
| 67 |
+
"loss": 30.4245,
|
| 68 |
+
"mean_token_accuracy": 0.6932173065841198,
|
| 69 |
+
"num_tokens": 1701193.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 0.8636196419596672,
|
| 74 |
+
"epoch": 0.112,
|
| 75 |
+
"grad_norm": 63.5,
|
| 76 |
+
"learning_rate": 4.600000000000001e-06,
|
| 77 |
+
"loss": 28.963,
|
| 78 |
+
"mean_token_accuracy": 0.6949520386755467,
|
| 79 |
+
"num_tokens": 1979232.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 0.986144457757473,
|
| 84 |
+
"epoch": 0.128,
|
| 85 |
+
"grad_norm": 45.25,
|
| 86 |
+
"learning_rate": 5.2666666666666665e-06,
|
| 87 |
+
"loss": 27.4106,
|
| 88 |
+
"mean_token_accuracy": 0.7027070496231318,
|
| 89 |
+
"num_tokens": 2274177.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.0647853959351778,
|
| 94 |
+
"epoch": 0.144,
|
| 95 |
+
"grad_norm": 34.5,
|
| 96 |
+
"learning_rate": 5.933333333333335e-06,
|
| 97 |
+
"loss": 25.7084,
|
| 98 |
+
"mean_token_accuracy": 0.7127731442451477,
|
| 99 |
+
"num_tokens": 2548445.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.129902769997716,
|
| 104 |
+
"epoch": 0.16,
|
| 105 |
+
"grad_norm": 29.625,
|
| 106 |
+
"learning_rate": 6.600000000000001e-06,
|
| 107 |
+
"loss": 23.6215,
|
| 108 |
+
"mean_token_accuracy": 0.724789521843195,
|
| 109 |
+
"num_tokens": 2824418.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.16,
|
| 114 |
+
"eval_biology_entropy": 3.7158153285980227,
|
| 115 |
+
"eval_biology_loss": 4.645914554595947,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.3934867687225342,
|
| 117 |
+
"eval_biology_num_tokens": 2824418.0,
|
| 118 |
+
"eval_biology_runtime": 58.324,
|
| 119 |
+
"eval_biology_samples_per_second": 8.573,
|
| 120 |
+
"eval_biology_steps_per_second": 2.143,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.16,
|
| 125 |
+
"eval_chemistry_entropy": 2.4825774974823,
|
| 126 |
+
"eval_chemistry_loss": 2.797250509262085,
|
| 127 |
+
"eval_chemistry_mean_token_accuracy": 0.5806102266311646,
|
| 128 |
+
"eval_chemistry_num_tokens": 2824418.0,
|
| 129 |
+
"eval_chemistry_runtime": 72.1529,
|
| 130 |
+
"eval_chemistry_samples_per_second": 6.93,
|
| 131 |
+
"eval_chemistry_steps_per_second": 1.732,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.16,
|
| 136 |
+
"eval_math_entropy": 1.2174424457550048,
|
| 137 |
+
"eval_math_loss": 1.4766321182250977,
|
| 138 |
+
"eval_math_mean_token_accuracy": 0.7193697824478149,
|
| 139 |
+
"eval_math_num_tokens": 2824418.0,
|
| 140 |
+
"eval_math_runtime": 75.2312,
|
| 141 |
+
"eval_math_samples_per_second": 6.646,
|
| 142 |
+
"eval_math_steps_per_second": 1.662,
|
| 143 |
+
"step": 100
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16,
|
| 147 |
+
"eval_physics_entropy": 2.0236835989952087,
|
| 148 |
+
"eval_physics_loss": 2.248775005340576,
|
| 149 |
+
"eval_physics_mean_token_accuracy": 0.6399151833057404,
|
| 150 |
+
"eval_physics_num_tokens": 2824418.0,
|
| 151 |
+
"eval_physics_runtime": 85.7293,
|
| 152 |
+
"eval_physics_samples_per_second": 5.832,
|
| 153 |
+
"eval_physics_steps_per_second": 1.458,
|
| 154 |
+
"step": 100
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"entropy": 1.1890651997178794,
|
| 158 |
+
"epoch": 0.176,
|
| 159 |
+
"grad_norm": 24.5,
|
| 160 |
+
"learning_rate": 7.266666666666668e-06,
|
| 161 |
+
"loss": 22.5161,
|
| 162 |
+
"mean_token_accuracy": 0.7291781000792981,
|
| 163 |
+
"num_tokens": 3110313.0,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"entropy": 1.2231741398572922,
|
| 168 |
+
"epoch": 0.192,
|
| 169 |
+
"grad_norm": 19.375,
|
| 170 |
+
"learning_rate": 7.933333333333334e-06,
|
| 171 |
+
"loss": 22.3058,
|
| 172 |
+
"mean_token_accuracy": 0.7288031376898289,
|
| 173 |
+
"num_tokens": 3394170.0,
|
| 174 |
+
"step": 120
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"entropy": 1.1990749202668667,
|
| 178 |
+
"epoch": 0.208,
|
| 179 |
+
"grad_norm": 17.125,
|
| 180 |
+
"learning_rate": 8.6e-06,
|
| 181 |
+
"loss": 21.6321,
|
| 182 |
+
"mean_token_accuracy": 0.7331194914877415,
|
| 183 |
+
"num_tokens": 3673600.0,
|
| 184 |
+
"step": 130
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"entropy": 1.1935963843017816,
|
| 188 |
+
"epoch": 0.224,
|
| 189 |
+
"grad_norm": 18.5,
|
| 190 |
+
"learning_rate": 9.266666666666667e-06,
|
| 191 |
+
"loss": 20.9973,
|
| 192 |
+
"mean_token_accuracy": 0.7374927394092083,
|
| 193 |
+
"num_tokens": 3953732.0,
|
| 194 |
+
"step": 140
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"entropy": 1.1516071777790784,
|
| 198 |
+
"epoch": 0.24,
|
| 199 |
+
"grad_norm": 15.6875,
|
| 200 |
+
"learning_rate": 9.933333333333334e-06,
|
| 201 |
+
"loss": 20.0443,
|
| 202 |
+
"mean_token_accuracy": 0.7471608892083168,
|
| 203 |
+
"num_tokens": 4243655.0,
|
| 204 |
+
"step": 150
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"entropy": 1.1274361703544855,
|
| 208 |
+
"epoch": 0.256,
|
| 209 |
+
"grad_norm": 17.125,
|
| 210 |
+
"learning_rate": 1.0600000000000002e-05,
|
| 211 |
+
"loss": 19.5101,
|
| 212 |
+
"mean_token_accuracy": 0.7516594093292952,
|
| 213 |
+
"num_tokens": 4531471.0,
|
| 214 |
+
"step": 160
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"entropy": 1.140014799311757,
|
| 218 |
+
"epoch": 0.272,
|
| 219 |
+
"grad_norm": 14.0,
|
| 220 |
+
"learning_rate": 1.1266666666666668e-05,
|
| 221 |
+
"loss": 19.5844,
|
| 222 |
+
"mean_token_accuracy": 0.7498023040592671,
|
| 223 |
+
"num_tokens": 4810284.0,
|
| 224 |
+
"step": 170
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"entropy": 1.1203772608190774,
|
| 228 |
+
"epoch": 0.288,
|
| 229 |
+
"grad_norm": 12.5625,
|
| 230 |
+
"learning_rate": 1.1933333333333335e-05,
|
| 231 |
+
"loss": 19.1823,
|
| 232 |
+
"mean_token_accuracy": 0.753832882642746,
|
| 233 |
+
"num_tokens": 5095104.0,
|
| 234 |
+
"step": 180
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"entropy": 1.129030792787671,
|
| 238 |
+
"epoch": 0.304,
|
| 239 |
+
"grad_norm": 12.9375,
|
| 240 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 241 |
+
"loss": 19.1862,
|
| 242 |
+
"mean_token_accuracy": 0.7524638257920742,
|
| 243 |
+
"num_tokens": 5383732.0,
|
| 244 |
+
"step": 190
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"entropy": 1.090552070736885,
|
| 248 |
+
"epoch": 0.32,
|
| 249 |
+
"grad_norm": 15.0625,
|
| 250 |
+
"learning_rate": 1.3266666666666668e-05,
|
| 251 |
+
"loss": 18.5807,
|
| 252 |
+
"mean_token_accuracy": 0.7593985345214606,
|
| 253 |
+
"num_tokens": 5676334.0,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.32,
|
| 258 |
+
"eval_biology_entropy": 3.458996726989746,
|
| 259 |
+
"eval_biology_loss": 4.269806385040283,
|
| 260 |
+
"eval_biology_mean_token_accuracy": 0.4222190487384796,
|
| 261 |
+
"eval_biology_num_tokens": 5676334.0,
|
| 262 |
+
"eval_biology_runtime": 58.3189,
|
| 263 |
+
"eval_biology_samples_per_second": 8.574,
|
| 264 |
+
"eval_biology_steps_per_second": 2.143,
|
| 265 |
+
"step": 200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.32,
|
| 269 |
+
"eval_chemistry_entropy": 2.180737380027771,
|
| 270 |
+
"eval_chemistry_loss": 2.439823865890503,
|
| 271 |
+
"eval_chemistry_mean_token_accuracy": 0.6122945117950439,
|
| 272 |
+
"eval_chemistry_num_tokens": 5676334.0,
|
| 273 |
+
"eval_chemistry_runtime": 72.4241,
|
| 274 |
+
"eval_chemistry_samples_per_second": 6.904,
|
| 275 |
+
"eval_chemistry_steps_per_second": 1.726,
|
| 276 |
+
"step": 200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.32,
|
| 280 |
+
"eval_math_entropy": 1.1109818544387817,
|
| 281 |
+
"eval_math_loss": 1.1917932033538818,
|
| 282 |
+
"eval_math_mean_token_accuracy": 0.752472719669342,
|
| 283 |
+
"eval_math_num_tokens": 5676334.0,
|
| 284 |
+
"eval_math_runtime": 75.8098,
|
| 285 |
+
"eval_math_samples_per_second": 6.595,
|
| 286 |
+
"eval_math_steps_per_second": 1.649,
|
| 287 |
+
"step": 200
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.32,
|
| 291 |
+
"eval_physics_entropy": 1.7737830457687378,
|
| 292 |
+
"eval_physics_loss": 1.935411810874939,
|
| 293 |
+
"eval_physics_mean_token_accuracy": 0.670736522436142,
|
| 294 |
+
"eval_physics_num_tokens": 5676334.0,
|
| 295 |
+
"eval_physics_runtime": 86.0358,
|
| 296 |
+
"eval_physics_samples_per_second": 5.812,
|
| 297 |
+
"eval_physics_steps_per_second": 1.453,
|
| 298 |
+
"step": 200
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"entropy": 1.1083258152008058,
|
| 302 |
+
"epoch": 0.336,
|
| 303 |
+
"grad_norm": 14.875,
|
| 304 |
+
"learning_rate": 1.3933333333333334e-05,
|
| 305 |
+
"loss": 18.9806,
|
| 306 |
+
"mean_token_accuracy": 0.7541178163141012,
|
| 307 |
+
"num_tokens": 5958480.0,
|
| 308 |
+
"step": 210
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"entropy": 1.0838440161198377,
|
| 312 |
+
"epoch": 0.352,
|
| 313 |
+
"grad_norm": 13.625,
|
| 314 |
+
"learning_rate": 1.46e-05,
|
| 315 |
+
"loss": 18.2469,
|
| 316 |
+
"mean_token_accuracy": 0.7600458469241858,
|
| 317 |
+
"num_tokens": 6242161.0,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"entropy": 1.0777738582342864,
|
| 322 |
+
"epoch": 0.368,
|
| 323 |
+
"grad_norm": 14.6875,
|
| 324 |
+
"learning_rate": 1.5266666666666667e-05,
|
| 325 |
+
"loss": 18.3345,
|
| 326 |
+
"mean_token_accuracy": 0.759134492278099,
|
| 327 |
+
"num_tokens": 6523679.0,
|
| 328 |
+
"step": 230
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"entropy": 1.0698408473283052,
|
| 332 |
+
"epoch": 0.384,
|
| 333 |
+
"grad_norm": 14.0,
|
| 334 |
+
"learning_rate": 1.5933333333333336e-05,
|
| 335 |
+
"loss": 17.9275,
|
| 336 |
+
"mean_token_accuracy": 0.7646602354943752,
|
| 337 |
+
"num_tokens": 6810978.0,
|
| 338 |
+
"step": 240
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"entropy": 1.0921947304159403,
|
| 342 |
+
"epoch": 0.4,
|
| 343 |
+
"grad_norm": 13.8125,
|
| 344 |
+
"learning_rate": 1.66e-05,
|
| 345 |
+
"loss": 18.5932,
|
| 346 |
+
"mean_token_accuracy": 0.757987079769373,
|
| 347 |
+
"num_tokens": 7096903.0,
|
| 348 |
+
"step": 250
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"entropy": 1.0657698467373848,
|
| 352 |
+
"epoch": 0.416,
|
| 353 |
+
"grad_norm": 12.9375,
|
| 354 |
+
"learning_rate": 1.726666666666667e-05,
|
| 355 |
+
"loss": 17.8874,
|
| 356 |
+
"mean_token_accuracy": 0.7630033176392317,
|
| 357 |
+
"num_tokens": 7377181.0,
|
| 358 |
+
"step": 260
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"entropy": 1.0416713248938323,
|
| 362 |
+
"epoch": 0.432,
|
| 363 |
+
"grad_norm": 15.1875,
|
| 364 |
+
"learning_rate": 1.7933333333333333e-05,
|
| 365 |
+
"loss": 17.5118,
|
| 366 |
+
"mean_token_accuracy": 0.7672814972698688,
|
| 367 |
+
"num_tokens": 7650523.0,
|
| 368 |
+
"step": 270
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"entropy": 1.0186962354928255,
|
| 372 |
+
"epoch": 0.448,
|
| 373 |
+
"grad_norm": 15.3125,
|
| 374 |
+
"learning_rate": 1.86e-05,
|
| 375 |
+
"loss": 17.3118,
|
| 376 |
+
"mean_token_accuracy": 0.7682009607553482,
|
| 377 |
+
"num_tokens": 7936788.0,
|
| 378 |
+
"step": 280
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"entropy": 1.036542361229658,
|
| 382 |
+
"epoch": 0.464,
|
| 383 |
+
"grad_norm": 15.8125,
|
| 384 |
+
"learning_rate": 1.926666666666667e-05,
|
| 385 |
+
"loss": 17.3748,
|
| 386 |
+
"mean_token_accuracy": 0.7694676879793405,
|
| 387 |
+
"num_tokens": 8223147.0,
|
| 388 |
+
"step": 290
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"entropy": 1.0569938763976097,
|
| 392 |
+
"epoch": 0.48,
|
| 393 |
+
"grad_norm": 14.5,
|
| 394 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 395 |
+
"loss": 17.6393,
|
| 396 |
+
"mean_token_accuracy": 0.7655092909932136,
|
| 397 |
+
"num_tokens": 8506030.0,
|
| 398 |
+
"step": 300
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.48,
|
| 402 |
+
"eval_biology_entropy": 3.567513717651367,
|
| 403 |
+
"eval_biology_loss": 4.103325366973877,
|
| 404 |
+
"eval_biology_mean_token_accuracy": 0.43289893221855164,
|
| 405 |
+
"eval_biology_num_tokens": 8506030.0,
|
| 406 |
+
"eval_biology_runtime": 58.2179,
|
| 407 |
+
"eval_biology_samples_per_second": 8.588,
|
| 408 |
+
"eval_biology_steps_per_second": 2.147,
|
| 409 |
+
"step": 300
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.48,
|
| 413 |
+
"eval_chemistry_entropy": 2.159853268146515,
|
| 414 |
+
"eval_chemistry_loss": 2.3111026287078857,
|
| 415 |
+
"eval_chemistry_mean_token_accuracy": 0.6211932191848755,
|
| 416 |
+
"eval_chemistry_num_tokens": 8506030.0,
|
| 417 |
+
"eval_chemistry_runtime": 72.4538,
|
| 418 |
+
"eval_chemistry_samples_per_second": 6.901,
|
| 419 |
+
"eval_chemistry_steps_per_second": 1.725,
|
| 420 |
+
"step": 300
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.48,
|
| 424 |
+
"eval_math_entropy": 1.055273208618164,
|
| 425 |
+
"eval_math_loss": 1.103938341140747,
|
| 426 |
+
"eval_math_mean_token_accuracy": 0.7635622372627259,
|
| 427 |
+
"eval_math_num_tokens": 8506030.0,
|
| 428 |
+
"eval_math_runtime": 75.7526,
|
| 429 |
+
"eval_math_samples_per_second": 6.6,
|
| 430 |
+
"eval_math_steps_per_second": 1.65,
|
| 431 |
+
"step": 300
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.48,
|
| 435 |
+
"eval_physics_entropy": 1.7374450120925904,
|
| 436 |
+
"eval_physics_loss": 1.8264347314834595,
|
| 437 |
+
"eval_physics_mean_token_accuracy": 0.6800054814815522,
|
| 438 |
+
"eval_physics_num_tokens": 8506030.0,
|
| 439 |
+
"eval_physics_runtime": 85.6984,
|
| 440 |
+
"eval_physics_samples_per_second": 5.834,
|
| 441 |
+
"eval_physics_steps_per_second": 1.459,
|
| 442 |
+
"step": 300
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"entropy": 1.0048395186662673,
|
| 446 |
+
"epoch": 0.496,
|
| 447 |
+
"grad_norm": 20.75,
|
| 448 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 449 |
+
"loss": 17.0439,
|
| 450 |
+
"mean_token_accuracy": 0.7706491403281689,
|
| 451 |
+
"num_tokens": 8788726.0,
|
| 452 |
+
"step": 310
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"entropy": 1.0206645488739015,
|
| 456 |
+
"epoch": 0.512,
|
| 457 |
+
"grad_norm": 14.125,
|
| 458 |
+
"learning_rate": 1.985925925925926e-05,
|
| 459 |
+
"loss": 17.0396,
|
| 460 |
+
"mean_token_accuracy": 0.7719013765454292,
|
| 461 |
+
"num_tokens": 9078039.0,
|
| 462 |
+
"step": 320
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"entropy": 0.9999576449394226,
|
| 466 |
+
"epoch": 0.528,
|
| 467 |
+
"grad_norm": 15.0625,
|
| 468 |
+
"learning_rate": 1.9785185185185187e-05,
|
| 469 |
+
"loss": 16.7994,
|
| 470 |
+
"mean_token_accuracy": 0.7749712273478508,
|
| 471 |
+
"num_tokens": 9373860.0,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"entropy": 1.0083586767315864,
|
| 476 |
+
"epoch": 0.544,
|
| 477 |
+
"grad_norm": 15.0625,
|
| 478 |
+
"learning_rate": 1.971111111111111e-05,
|
| 479 |
+
"loss": 16.9209,
|
| 480 |
+
"mean_token_accuracy": 0.7731953993439674,
|
| 481 |
+
"num_tokens": 9660940.0,
|
| 482 |
+
"step": 340
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"entropy": 1.0532048616558314,
|
| 486 |
+
"epoch": 0.56,
|
| 487 |
+
"grad_norm": 18.75,
|
| 488 |
+
"learning_rate": 1.963703703703704e-05,
|
| 489 |
+
"loss": 17.638,
|
| 490 |
+
"mean_token_accuracy": 0.7637613207101822,
|
| 491 |
+
"num_tokens": 9932302.0,
|
| 492 |
+
"step": 350
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"entropy": 1.0087582137435676,
|
| 496 |
+
"epoch": 0.576,
|
| 497 |
+
"grad_norm": 17.75,
|
| 498 |
+
"learning_rate": 1.9562962962962964e-05,
|
| 499 |
+
"loss": 17.0002,
|
| 500 |
+
"mean_token_accuracy": 0.7717114143073559,
|
| 501 |
+
"num_tokens": 10215462.0,
|
| 502 |
+
"step": 360
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"entropy": 1.0042562756687403,
|
| 506 |
+
"epoch": 0.592,
|
| 507 |
+
"grad_norm": 15.0,
|
| 508 |
+
"learning_rate": 1.948888888888889e-05,
|
| 509 |
+
"loss": 16.7266,
|
| 510 |
+
"mean_token_accuracy": 0.7740717619657517,
|
| 511 |
+
"num_tokens": 10504396.0,
|
| 512 |
+
"step": 370
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"entropy": 0.9980081049725413,
|
| 516 |
+
"epoch": 0.608,
|
| 517 |
+
"grad_norm": 16.125,
|
| 518 |
+
"learning_rate": 1.9414814814814817e-05,
|
| 519 |
+
"loss": 16.8411,
|
| 520 |
+
"mean_token_accuracy": 0.7726139184087515,
|
| 521 |
+
"num_tokens": 10793126.0,
|
| 522 |
+
"step": 380
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"entropy": 0.9865465115755796,
|
| 526 |
+
"epoch": 0.624,
|
| 527 |
+
"grad_norm": 14.5,
|
| 528 |
+
"learning_rate": 1.9340740740740743e-05,
|
| 529 |
+
"loss": 16.5077,
|
| 530 |
+
"mean_token_accuracy": 0.7759908959269524,
|
| 531 |
+
"num_tokens": 11081768.0,
|
| 532 |
+
"step": 390
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"entropy": 0.9753864344209433,
|
| 536 |
+
"epoch": 0.64,
|
| 537 |
+
"grad_norm": 13.0625,
|
| 538 |
+
"learning_rate": 1.926666666666667e-05,
|
| 539 |
+
"loss": 16.3328,
|
| 540 |
+
"mean_token_accuracy": 0.777353797480464,
|
| 541 |
+
"num_tokens": 11370320.0,
|
| 542 |
+
"step": 400
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.64,
|
| 546 |
+
"eval_biology_entropy": 3.4534600849151613,
|
| 547 |
+
"eval_biology_loss": 4.02693510055542,
|
| 548 |
+
"eval_biology_mean_token_accuracy": 0.4380790627002716,
|
| 549 |
+
"eval_biology_num_tokens": 11370320.0,
|
| 550 |
+
"eval_biology_runtime": 58.2953,
|
| 551 |
+
"eval_biology_samples_per_second": 8.577,
|
| 552 |
+
"eval_biology_steps_per_second": 2.144,
|
| 553 |
+
"step": 400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.64,
|
| 557 |
+
"eval_chemistry_entropy": 2.068386361122131,
|
| 558 |
+
"eval_chemistry_loss": 2.252868175506592,
|
| 559 |
+
"eval_chemistry_mean_token_accuracy": 0.6284125361442566,
|
| 560 |
+
"eval_chemistry_num_tokens": 11370320.0,
|
| 561 |
+
"eval_chemistry_runtime": 72.4168,
|
| 562 |
+
"eval_chemistry_samples_per_second": 6.904,
|
| 563 |
+
"eval_chemistry_steps_per_second": 1.726,
|
| 564 |
+
"step": 400
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.64,
|
| 568 |
+
"eval_math_entropy": 1.008522126674652,
|
| 569 |
+
"eval_math_loss": 1.0611059665679932,
|
| 570 |
+
"eval_math_mean_token_accuracy": 0.7697986388206481,
|
| 571 |
+
"eval_math_num_tokens": 11370320.0,
|
| 572 |
+
"eval_math_runtime": 75.8656,
|
| 573 |
+
"eval_math_samples_per_second": 6.591,
|
| 574 |
+
"eval_math_steps_per_second": 1.648,
|
| 575 |
+
"step": 400
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.64,
|
| 579 |
+
"eval_physics_entropy": 1.6607081370353698,
|
| 580 |
+
"eval_physics_loss": 1.7777302265167236,
|
| 581 |
+
"eval_physics_mean_token_accuracy": 0.685634191274643,
|
| 582 |
+
"eval_physics_num_tokens": 11370320.0,
|
| 583 |
+
"eval_physics_runtime": 86.0729,
|
| 584 |
+
"eval_physics_samples_per_second": 5.809,
|
| 585 |
+
"eval_physics_steps_per_second": 1.452,
|
| 586 |
+
"step": 400
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"entropy": 0.9546368110924959,
|
| 590 |
+
"epoch": 0.656,
|
| 591 |
+
"grad_norm": 13.3125,
|
| 592 |
+
"learning_rate": 1.9192592592592593e-05,
|
| 593 |
+
"loss": 16.1585,
|
| 594 |
+
"mean_token_accuracy": 0.7791520744562149,
|
| 595 |
+
"num_tokens": 11657835.0,
|
| 596 |
+
"step": 410
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"entropy": 1.0121946074068546,
|
| 600 |
+
"epoch": 0.672,
|
| 601 |
+
"grad_norm": 14.9375,
|
| 602 |
+
"learning_rate": 1.911851851851852e-05,
|
| 603 |
+
"loss": 16.7638,
|
| 604 |
+
"mean_token_accuracy": 0.7723276257514954,
|
| 605 |
+
"num_tokens": 11949262.0,
|
| 606 |
+
"step": 420
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"entropy": 0.9564321564510465,
|
| 610 |
+
"epoch": 0.688,
|
| 611 |
+
"grad_norm": 12.875,
|
| 612 |
+
"learning_rate": 1.9044444444444446e-05,
|
| 613 |
+
"loss": 16.2106,
|
| 614 |
+
"mean_token_accuracy": 0.7782567508518696,
|
| 615 |
+
"num_tokens": 12227640.0,
|
| 616 |
+
"step": 430
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"entropy": 0.9711398147046566,
|
| 620 |
+
"epoch": 0.704,
|
| 621 |
+
"grad_norm": 12.875,
|
| 622 |
+
"learning_rate": 1.8970370370370372e-05,
|
| 623 |
+
"loss": 16.2595,
|
| 624 |
+
"mean_token_accuracy": 0.7794518828392029,
|
| 625 |
+
"num_tokens": 12516641.0,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"entropy": 0.990043681114912,
|
| 630 |
+
"epoch": 0.72,
|
| 631 |
+
"grad_norm": 16.0,
|
| 632 |
+
"learning_rate": 1.8896296296296295e-05,
|
| 633 |
+
"loss": 16.6764,
|
| 634 |
+
"mean_token_accuracy": 0.7741847857832909,
|
| 635 |
+
"num_tokens": 12793343.0,
|
| 636 |
+
"step": 450
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"entropy": 0.9794478964060545,
|
| 640 |
+
"epoch": 0.736,
|
| 641 |
+
"grad_norm": 13.5625,
|
| 642 |
+
"learning_rate": 1.8822222222222225e-05,
|
| 643 |
+
"loss": 16.3042,
|
| 644 |
+
"mean_token_accuracy": 0.7780686043202877,
|
| 645 |
+
"num_tokens": 13077981.0,
|
| 646 |
+
"step": 460
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"entropy": 0.9613467682152986,
|
| 650 |
+
"epoch": 0.752,
|
| 651 |
+
"grad_norm": 14.6875,
|
| 652 |
+
"learning_rate": 1.874814814814815e-05,
|
| 653 |
+
"loss": 16.0345,
|
| 654 |
+
"mean_token_accuracy": 0.7808727387338876,
|
| 655 |
+
"num_tokens": 13358957.0,
|
| 656 |
+
"step": 470
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"entropy": 0.9237693391740323,
|
| 660 |
+
"epoch": 0.768,
|
| 661 |
+
"grad_norm": 12.0,
|
| 662 |
+
"learning_rate": 1.8674074074074075e-05,
|
| 663 |
+
"loss": 15.6444,
|
| 664 |
+
"mean_token_accuracy": 0.7876643743366003,
|
| 665 |
+
"num_tokens": 13653412.0,
|
| 666 |
+
"step": 480
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"entropy": 1.0019762709736824,
|
| 670 |
+
"epoch": 0.784,
|
| 671 |
+
"grad_norm": 15.6875,
|
| 672 |
+
"learning_rate": 1.86e-05,
|
| 673 |
+
"loss": 16.7034,
|
| 674 |
+
"mean_token_accuracy": 0.7748573537915945,
|
| 675 |
+
"num_tokens": 13940856.0,
|
| 676 |
+
"step": 490
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"entropy": 0.9975844457745552,
|
| 680 |
+
"epoch": 0.8,
|
| 681 |
+
"grad_norm": 17.0,
|
| 682 |
+
"learning_rate": 1.8525925925925928e-05,
|
| 683 |
+
"loss": 16.7128,
|
| 684 |
+
"mean_token_accuracy": 0.77371263243258,
|
| 685 |
+
"num_tokens": 14230754.0,
|
| 686 |
+
"step": 500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.8,
|
| 690 |
+
"eval_biology_entropy": 3.498454050064087,
|
| 691 |
+
"eval_biology_loss": 3.949568271636963,
|
| 692 |
+
"eval_biology_mean_token_accuracy": 0.44270548820495603,
|
| 693 |
+
"eval_biology_num_tokens": 14230754.0,
|
| 694 |
+
"eval_biology_runtime": 58.3602,
|
| 695 |
+
"eval_biology_samples_per_second": 8.567,
|
| 696 |
+
"eval_biology_steps_per_second": 2.142,
|
| 697 |
+
"step": 500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.8,
|
| 701 |
+
"eval_chemistry_entropy": 2.073633497238159,
|
| 702 |
+
"eval_chemistry_loss": 2.1966874599456787,
|
| 703 |
+
"eval_chemistry_mean_token_accuracy": 0.6324296245574951,
|
| 704 |
+
"eval_chemistry_num_tokens": 14230754.0,
|
| 705 |
+
"eval_chemistry_runtime": 71.3745,
|
| 706 |
+
"eval_chemistry_samples_per_second": 7.005,
|
| 707 |
+
"eval_chemistry_steps_per_second": 1.751,
|
| 708 |
+
"step": 500
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.8,
|
| 712 |
+
"eval_math_entropy": 0.9948915004730224,
|
| 713 |
+
"eval_math_loss": 1.0365564823150635,
|
| 714 |
+
"eval_math_mean_token_accuracy": 0.7735790548324585,
|
| 715 |
+
"eval_math_num_tokens": 14230754.0,
|
| 716 |
+
"eval_math_runtime": 75.2012,
|
| 717 |
+
"eval_math_samples_per_second": 6.649,
|
| 718 |
+
"eval_math_steps_per_second": 1.662,
|
| 719 |
+
"step": 500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.8,
|
| 723 |
+
"eval_physics_entropy": 1.6648971457481385,
|
| 724 |
+
"eval_physics_loss": 1.736081838607788,
|
| 725 |
+
"eval_physics_mean_token_accuracy": 0.6900339169502259,
|
| 726 |
+
"eval_physics_num_tokens": 14230754.0,
|
| 727 |
+
"eval_physics_runtime": 84.9467,
|
| 728 |
+
"eval_physics_samples_per_second": 5.886,
|
| 729 |
+
"eval_physics_steps_per_second": 1.472,
|
| 730 |
+
"step": 500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 0.9665802534669637,
|
| 734 |
+
"epoch": 0.816,
|
| 735 |
+
"grad_norm": 15.5,
|
| 736 |
+
"learning_rate": 1.8451851851851855e-05,
|
| 737 |
+
"loss": 16.0969,
|
| 738 |
+
"mean_token_accuracy": 0.77869370020926,
|
| 739 |
+
"num_tokens": 14519893.0,
|
| 740 |
+
"step": 510
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 0.9573450323194266,
|
| 744 |
+
"epoch": 0.832,
|
| 745 |
+
"grad_norm": 14.625,
|
| 746 |
+
"learning_rate": 1.8377777777777778e-05,
|
| 747 |
+
"loss": 16.1991,
|
| 748 |
+
"mean_token_accuracy": 0.7803513094782829,
|
| 749 |
+
"num_tokens": 14805088.0,
|
| 750 |
+
"step": 520
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 0.9879640638828278,
|
| 754 |
+
"epoch": 0.848,
|
| 755 |
+
"grad_norm": 14.0625,
|
| 756 |
+
"learning_rate": 1.8303703703703704e-05,
|
| 757 |
+
"loss": 16.5646,
|
| 758 |
+
"mean_token_accuracy": 0.7745878294110298,
|
| 759 |
+
"num_tokens": 15086306.0,
|
| 760 |
+
"step": 530
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 0.974746185541153,
|
| 764 |
+
"epoch": 0.864,
|
| 765 |
+
"grad_norm": 13.6875,
|
| 766 |
+
"learning_rate": 1.822962962962963e-05,
|
| 767 |
+
"loss": 16.3173,
|
| 768 |
+
"mean_token_accuracy": 0.7770581971853971,
|
| 769 |
+
"num_tokens": 15370985.0,
|
| 770 |
+
"step": 540
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 0.9910150848329067,
|
| 774 |
+
"epoch": 0.88,
|
| 775 |
+
"grad_norm": 17.0,
|
| 776 |
+
"learning_rate": 1.8155555555555557e-05,
|
| 777 |
+
"loss": 16.5347,
|
| 778 |
+
"mean_token_accuracy": 0.7754124801605939,
|
| 779 |
+
"num_tokens": 15653836.0,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 0.9755746092647314,
|
| 784 |
+
"epoch": 0.896,
|
| 785 |
+
"grad_norm": 14.3125,
|
| 786 |
+
"learning_rate": 1.8081481481481484e-05,
|
| 787 |
+
"loss": 16.3696,
|
| 788 |
+
"mean_token_accuracy": 0.7775021918118,
|
| 789 |
+
"num_tokens": 15932179.0,
|
| 790 |
+
"step": 560
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 0.948325376585126,
|
| 794 |
+
"epoch": 0.912,
|
| 795 |
+
"grad_norm": 14.8125,
|
| 796 |
+
"learning_rate": 1.800740740740741e-05,
|
| 797 |
+
"loss": 15.8592,
|
| 798 |
+
"mean_token_accuracy": 0.7806785147637129,
|
| 799 |
+
"num_tokens": 16219842.0,
|
| 800 |
+
"step": 570
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.9481973759829998,
|
| 804 |
+
"epoch": 0.928,
|
| 805 |
+
"grad_norm": 14.8125,
|
| 806 |
+
"learning_rate": 1.7933333333333333e-05,
|
| 807 |
+
"loss": 15.8814,
|
| 808 |
+
"mean_token_accuracy": 0.782349206879735,
|
| 809 |
+
"num_tokens": 16499842.0,
|
| 810 |
+
"step": 580
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 0.9722379047423602,
|
| 814 |
+
"epoch": 0.944,
|
| 815 |
+
"grad_norm": 15.375,
|
| 816 |
+
"learning_rate": 1.785925925925926e-05,
|
| 817 |
+
"loss": 16.4337,
|
| 818 |
+
"mean_token_accuracy": 0.776747428253293,
|
| 819 |
+
"num_tokens": 16781882.0,
|
| 820 |
+
"step": 590
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 0.9909376438707114,
|
| 824 |
+
"epoch": 0.96,
|
| 825 |
+
"grad_norm": 14.6875,
|
| 826 |
+
"learning_rate": 1.7785185185185186e-05,
|
| 827 |
+
"loss": 16.4215,
|
| 828 |
+
"mean_token_accuracy": 0.7778498761355876,
|
| 829 |
+
"num_tokens": 17067407.0,
|
| 830 |
+
"step": 600
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.96,
|
| 834 |
+
"eval_biology_entropy": 3.4036583194732666,
|
| 835 |
+
"eval_biology_loss": 3.9303975105285645,
|
| 836 |
+
"eval_biology_mean_token_accuracy": 0.4453940885066986,
|
| 837 |
+
"eval_biology_num_tokens": 17067407.0,
|
| 838 |
+
"eval_biology_runtime": 58.2933,
|
| 839 |
+
"eval_biology_samples_per_second": 8.577,
|
| 840 |
+
"eval_biology_steps_per_second": 2.144,
|
| 841 |
+
"step": 600
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.96,
|
| 845 |
+
"eval_chemistry_entropy": 2.0066872639656066,
|
| 846 |
+
"eval_chemistry_loss": 2.172046422958374,
|
| 847 |
+
"eval_chemistry_mean_token_accuracy": 0.635353732585907,
|
| 848 |
+
"eval_chemistry_num_tokens": 17067407.0,
|
| 849 |
+
"eval_chemistry_runtime": 72.0916,
|
| 850 |
+
"eval_chemistry_samples_per_second": 6.936,
|
| 851 |
+
"eval_chemistry_steps_per_second": 1.734,
|
| 852 |
+
"step": 600
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.96,
|
| 856 |
+
"eval_math_entropy": 0.9714911403656006,
|
| 857 |
+
"eval_math_loss": 1.020026445388794,
|
| 858 |
+
"eval_math_mean_token_accuracy": 0.7759292826652527,
|
| 859 |
+
"eval_math_num_tokens": 17067407.0,
|
| 860 |
+
"eval_math_runtime": 75.4884,
|
| 861 |
+
"eval_math_samples_per_second": 6.624,
|
| 862 |
+
"eval_math_steps_per_second": 1.656,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.96,
|
| 867 |
+
"eval_physics_entropy": 1.613628080844879,
|
| 868 |
+
"eval_physics_loss": 1.7234340906143188,
|
| 869 |
+
"eval_physics_mean_token_accuracy": 0.6920417103767394,
|
| 870 |
+
"eval_physics_num_tokens": 17067407.0,
|
| 871 |
+
"eval_physics_runtime": 85.693,
|
| 872 |
+
"eval_physics_samples_per_second": 5.835,
|
| 873 |
+
"eval_physics_steps_per_second": 1.459,
|
| 874 |
+
"step": 600
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"entropy": 0.9597863841801881,
|
| 878 |
+
"epoch": 0.976,
|
| 879 |
+
"grad_norm": 20.375,
|
| 880 |
+
"learning_rate": 1.7711111111111113e-05,
|
| 881 |
+
"loss": 16.0384,
|
| 882 |
+
"mean_token_accuracy": 0.7816183552145958,
|
| 883 |
+
"num_tokens": 17350994.0,
|
| 884 |
+
"step": 610
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"entropy": 0.973047361895442,
|
| 888 |
+
"epoch": 0.992,
|
| 889 |
+
"grad_norm": 13.9375,
|
| 890 |
+
"learning_rate": 1.763703703703704e-05,
|
| 891 |
+
"loss": 16.3168,
|
| 892 |
+
"mean_token_accuracy": 0.7756219923496246,
|
| 893 |
+
"num_tokens": 17637514.0,
|
| 894 |
+
"step": 620
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"entropy": 0.9593736134469509,
|
| 898 |
+
"epoch": 1.008,
|
| 899 |
+
"grad_norm": 12.75,
|
| 900 |
+
"learning_rate": 1.7562962962962962e-05,
|
| 901 |
+
"loss": 15.9379,
|
| 902 |
+
"mean_token_accuracy": 0.7799578707665205,
|
| 903 |
+
"num_tokens": 17926570.0,
|
| 904 |
+
"step": 630
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"entropy": 0.9494194338098169,
|
| 908 |
+
"epoch": 1.024,
|
| 909 |
+
"grad_norm": 14.125,
|
| 910 |
+
"learning_rate": 1.7488888888888892e-05,
|
| 911 |
+
"loss": 15.8718,
|
| 912 |
+
"mean_token_accuracy": 0.7815670046955348,
|
| 913 |
+
"num_tokens": 18207652.0,
|
| 914 |
+
"step": 640
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"entropy": 0.9487265404313803,
|
| 918 |
+
"epoch": 1.04,
|
| 919 |
+
"grad_norm": 16.125,
|
| 920 |
+
"learning_rate": 1.7414814814814815e-05,
|
| 921 |
+
"loss": 15.8653,
|
| 922 |
+
"mean_token_accuracy": 0.783008698746562,
|
| 923 |
+
"num_tokens": 18484931.0,
|
| 924 |
+
"step": 650
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"entropy": 0.9464438438415528,
|
| 928 |
+
"epoch": 1.056,
|
| 929 |
+
"grad_norm": 19.875,
|
| 930 |
+
"learning_rate": 1.7340740740740742e-05,
|
| 931 |
+
"loss": 15.9465,
|
| 932 |
+
"mean_token_accuracy": 0.7813190460205078,
|
| 933 |
+
"num_tokens": 18773457.0,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"entropy": 0.9594576228410006,
|
| 938 |
+
"epoch": 1.072,
|
| 939 |
+
"grad_norm": 15.0625,
|
| 940 |
+
"learning_rate": 1.726666666666667e-05,
|
| 941 |
+
"loss": 15.9518,
|
| 942 |
+
"mean_token_accuracy": 0.779586423188448,
|
| 943 |
+
"num_tokens": 19055365.0,
|
| 944 |
+
"step": 670
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"entropy": 0.9468420147895813,
|
| 948 |
+
"epoch": 1.088,
|
| 949 |
+
"grad_norm": 12.8125,
|
| 950 |
+
"learning_rate": 1.7192592592592595e-05,
|
| 951 |
+
"loss": 15.7798,
|
| 952 |
+
"mean_token_accuracy": 0.782842144370079,
|
| 953 |
+
"num_tokens": 19345730.0,
|
| 954 |
+
"step": 680
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"entropy": 0.9552154852077365,
|
| 958 |
+
"epoch": 1.104,
|
| 959 |
+
"grad_norm": 12.0625,
|
| 960 |
+
"learning_rate": 1.711851851851852e-05,
|
| 961 |
+
"loss": 15.8369,
|
| 962 |
+
"mean_token_accuracy": 0.7829934328794479,
|
| 963 |
+
"num_tokens": 19637390.0,
|
| 964 |
+
"step": 690
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"entropy": 0.9152646586298943,
|
| 968 |
+
"epoch": 1.12,
|
| 969 |
+
"grad_norm": 16.125,
|
| 970 |
+
"learning_rate": 1.7044444444444445e-05,
|
| 971 |
+
"loss": 15.427,
|
| 972 |
+
"mean_token_accuracy": 0.7873255901038647,
|
| 973 |
+
"num_tokens": 19923914.0,
|
| 974 |
+
"step": 700
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.12,
|
| 978 |
+
"eval_biology_entropy": 3.324676399230957,
|
| 979 |
+
"eval_biology_loss": 3.911534070968628,
|
| 980 |
+
"eval_biology_mean_token_accuracy": 0.44601045060157773,
|
| 981 |
+
"eval_biology_num_tokens": 19923914.0,
|
| 982 |
+
"eval_biology_runtime": 58.5206,
|
| 983 |
+
"eval_biology_samples_per_second": 8.544,
|
| 984 |
+
"eval_biology_steps_per_second": 2.136,
|
| 985 |
+
"step": 700
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 1.12,
|
| 989 |
+
"eval_chemistry_entropy": 1.9495020627975463,
|
| 990 |
+
"eval_chemistry_loss": 2.1536266803741455,
|
| 991 |
+
"eval_chemistry_mean_token_accuracy": 0.6368214721679688,
|
| 992 |
+
"eval_chemistry_num_tokens": 19923914.0,
|
| 993 |
+
"eval_chemistry_runtime": 72.1357,
|
| 994 |
+
"eval_chemistry_samples_per_second": 6.931,
|
| 995 |
+
"eval_chemistry_steps_per_second": 1.733,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.12,
|
| 1000 |
+
"eval_math_entropy": 0.9499740815162658,
|
| 1001 |
+
"eval_math_loss": 1.0100210905075073,
|
| 1002 |
+
"eval_math_mean_token_accuracy": 0.7775792751312256,
|
| 1003 |
+
"eval_math_num_tokens": 19923914.0,
|
| 1004 |
+
"eval_math_runtime": 75.4806,
|
| 1005 |
+
"eval_math_samples_per_second": 6.624,
|
| 1006 |
+
"eval_math_steps_per_second": 1.656,
|
| 1007 |
+
"step": 700
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 1.12,
|
| 1011 |
+
"eval_physics_entropy": 1.5690358576774597,
|
| 1012 |
+
"eval_physics_loss": 1.7045847177505493,
|
| 1013 |
+
"eval_physics_mean_token_accuracy": 0.6933669624328613,
|
| 1014 |
+
"eval_physics_num_tokens": 19923914.0,
|
| 1015 |
+
"eval_physics_runtime": 85.7866,
|
| 1016 |
+
"eval_physics_samples_per_second": 5.828,
|
| 1017 |
+
"eval_physics_steps_per_second": 1.457,
|
| 1018 |
+
"step": 700
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"entropy": 0.9291105344891548,
|
| 1022 |
+
"epoch": 1.1360000000000001,
|
| 1023 |
+
"grad_norm": 14.4375,
|
| 1024 |
+
"learning_rate": 1.697037037037037e-05,
|
| 1025 |
+
"loss": 15.4986,
|
| 1026 |
+
"mean_token_accuracy": 0.7853703975677491,
|
| 1027 |
+
"num_tokens": 20201892.0,
|
| 1028 |
+
"step": 710
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"entropy": 0.9415947886183857,
|
| 1032 |
+
"epoch": 1.152,
|
| 1033 |
+
"grad_norm": 17.375,
|
| 1034 |
+
"learning_rate": 1.6896296296296298e-05,
|
| 1035 |
+
"loss": 15.7175,
|
| 1036 |
+
"mean_token_accuracy": 0.7841294795274735,
|
| 1037 |
+
"num_tokens": 20490282.0,
|
| 1038 |
+
"step": 720
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"entropy": 0.9340157110244036,
|
| 1042 |
+
"epoch": 1.168,
|
| 1043 |
+
"grad_norm": 13.625,
|
| 1044 |
+
"learning_rate": 1.6822222222222224e-05,
|
| 1045 |
+
"loss": 15.6449,
|
| 1046 |
+
"mean_token_accuracy": 0.7852166913449764,
|
| 1047 |
+
"num_tokens": 20785786.0,
|
| 1048 |
+
"step": 730
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"entropy": 0.9397943781688809,
|
| 1052 |
+
"epoch": 1.184,
|
| 1053 |
+
"grad_norm": 15.75,
|
| 1054 |
+
"learning_rate": 1.6748148148148147e-05,
|
| 1055 |
+
"loss": 15.6534,
|
| 1056 |
+
"mean_token_accuracy": 0.7839406657963991,
|
| 1057 |
+
"num_tokens": 21074205.0,
|
| 1058 |
+
"step": 740
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"entropy": 0.9129594217985868,
|
| 1062 |
+
"epoch": 1.2,
|
| 1063 |
+
"grad_norm": 17.375,
|
| 1064 |
+
"learning_rate": 1.6674074074074077e-05,
|
| 1065 |
+
"loss": 15.3952,
|
| 1066 |
+
"mean_token_accuracy": 0.7864840351045131,
|
| 1067 |
+
"num_tokens": 21369159.0,
|
| 1068 |
+
"step": 750
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"entropy": 0.9617549370974302,
|
| 1072 |
+
"epoch": 1.216,
|
| 1073 |
+
"grad_norm": 16.0,
|
| 1074 |
+
"learning_rate": 1.66e-05,
|
| 1075 |
+
"loss": 15.9238,
|
| 1076 |
+
"mean_token_accuracy": 0.7814570065587759,
|
| 1077 |
+
"num_tokens": 21649178.0,
|
| 1078 |
+
"step": 760
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"entropy": 0.916939202696085,
|
| 1082 |
+
"epoch": 1.232,
|
| 1083 |
+
"grad_norm": 15.5,
|
| 1084 |
+
"learning_rate": 1.6525925925925927e-05,
|
| 1085 |
+
"loss": 15.3019,
|
| 1086 |
+
"mean_token_accuracy": 0.788709907606244,
|
| 1087 |
+
"num_tokens": 21930239.0,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"entropy": 0.9114697994664311,
|
| 1092 |
+
"epoch": 1.248,
|
| 1093 |
+
"grad_norm": 14.6875,
|
| 1094 |
+
"learning_rate": 1.6451851851851853e-05,
|
| 1095 |
+
"loss": 15.1832,
|
| 1096 |
+
"mean_token_accuracy": 0.7884270831942558,
|
| 1097 |
+
"num_tokens": 22216387.0,
|
| 1098 |
+
"step": 780
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"entropy": 0.9520778905600309,
|
| 1102 |
+
"epoch": 1.264,
|
| 1103 |
+
"grad_norm": 14.9375,
|
| 1104 |
+
"learning_rate": 1.637777777777778e-05,
|
| 1105 |
+
"loss": 15.9046,
|
| 1106 |
+
"mean_token_accuracy": 0.7815822076052428,
|
| 1107 |
+
"num_tokens": 22501002.0,
|
| 1108 |
+
"step": 790
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"entropy": 0.9420083146542311,
|
| 1112 |
+
"epoch": 1.28,
|
| 1113 |
+
"grad_norm": 20.375,
|
| 1114 |
+
"learning_rate": 1.6303703703703706e-05,
|
| 1115 |
+
"loss": 15.8436,
|
| 1116 |
+
"mean_token_accuracy": 0.7832419756799937,
|
| 1117 |
+
"num_tokens": 22779682.0,
|
| 1118 |
+
"step": 800
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.28,
|
| 1122 |
+
"eval_biology_entropy": 3.384992992401123,
|
| 1123 |
+
"eval_biology_loss": 3.905174493789673,
|
| 1124 |
+
"eval_biology_mean_token_accuracy": 0.447558299779892,
|
| 1125 |
+
"eval_biology_num_tokens": 22779682.0,
|
| 1126 |
+
"eval_biology_runtime": 58.5193,
|
| 1127 |
+
"eval_biology_samples_per_second": 8.544,
|
| 1128 |
+
"eval_biology_steps_per_second": 2.136,
|
| 1129 |
+
"step": 800
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.28,
|
| 1133 |
+
"eval_chemistry_entropy": 1.9894180879592895,
|
| 1134 |
+
"eval_chemistry_loss": 2.1436502933502197,
|
| 1135 |
+
"eval_chemistry_mean_token_accuracy": 0.6379990315437317,
|
| 1136 |
+
"eval_chemistry_num_tokens": 22779682.0,
|
| 1137 |
+
"eval_chemistry_runtime": 72.08,
|
| 1138 |
+
"eval_chemistry_samples_per_second": 6.937,
|
| 1139 |
+
"eval_chemistry_steps_per_second": 1.734,
|
| 1140 |
+
"step": 800
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.28,
|
| 1144 |
+
"eval_math_entropy": 0.9536066613197327,
|
| 1145 |
+
"eval_math_loss": 1.0001386404037476,
|
| 1146 |
+
"eval_math_mean_token_accuracy": 0.7793465681076049,
|
| 1147 |
+
"eval_math_num_tokens": 22779682.0,
|
| 1148 |
+
"eval_math_runtime": 75.2623,
|
| 1149 |
+
"eval_math_samples_per_second": 6.643,
|
| 1150 |
+
"eval_math_steps_per_second": 1.661,
|
| 1151 |
+
"step": 800
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.28,
|
| 1155 |
+
"eval_physics_entropy": 1.6050884528160094,
|
| 1156 |
+
"eval_physics_loss": 1.695229172706604,
|
| 1157 |
+
"eval_physics_mean_token_accuracy": 0.6948762707710266,
|
| 1158 |
+
"eval_physics_num_tokens": 22779682.0,
|
| 1159 |
+
"eval_physics_runtime": 85.7894,
|
| 1160 |
+
"eval_physics_samples_per_second": 5.828,
|
| 1161 |
+
"eval_physics_steps_per_second": 1.457,
|
| 1162 |
+
"step": 800
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"entropy": 0.9270045252516865,
|
| 1166 |
+
"epoch": 1.296,
|
| 1167 |
+
"grad_norm": 15.0625,
|
| 1168 |
+
"learning_rate": 1.622962962962963e-05,
|
| 1169 |
+
"loss": 15.5761,
|
| 1170 |
+
"mean_token_accuracy": 0.7847258739173413,
|
| 1171 |
+
"num_tokens": 23057744.0,
|
| 1172 |
+
"step": 810
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"entropy": 0.9379040151834488,
|
| 1176 |
+
"epoch": 1.312,
|
| 1177 |
+
"grad_norm": 16.0,
|
| 1178 |
+
"learning_rate": 1.6155555555555556e-05,
|
| 1179 |
+
"loss": 15.6785,
|
| 1180 |
+
"mean_token_accuracy": 0.7841929640620947,
|
| 1181 |
+
"num_tokens": 23344644.0,
|
| 1182 |
+
"step": 820
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"entropy": 0.9579077050089836,
|
| 1186 |
+
"epoch": 1.328,
|
| 1187 |
+
"grad_norm": 18.75,
|
| 1188 |
+
"learning_rate": 1.6081481481481482e-05,
|
| 1189 |
+
"loss": 15.9331,
|
| 1190 |
+
"mean_token_accuracy": 0.7809579152613878,
|
| 1191 |
+
"num_tokens": 23622405.0,
|
| 1192 |
+
"step": 830
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"entropy": 0.9214272113516927,
|
| 1196 |
+
"epoch": 1.3439999999999999,
|
| 1197 |
+
"grad_norm": 14.3125,
|
| 1198 |
+
"learning_rate": 1.600740740740741e-05,
|
| 1199 |
+
"loss": 15.4675,
|
| 1200 |
+
"mean_token_accuracy": 0.7852523807436228,
|
| 1201 |
+
"num_tokens": 23899771.0,
|
| 1202 |
+
"step": 840
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"entropy": 0.9227435002103448,
|
| 1206 |
+
"epoch": 1.3599999999999999,
|
| 1207 |
+
"grad_norm": 16.5,
|
| 1208 |
+
"learning_rate": 1.5933333333333336e-05,
|
| 1209 |
+
"loss": 15.4552,
|
| 1210 |
+
"mean_token_accuracy": 0.7862110409885645,
|
| 1211 |
+
"num_tokens": 24187023.0,
|
| 1212 |
+
"step": 850
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"entropy": 0.9958576496690512,
|
| 1216 |
+
"epoch": 1.376,
|
| 1217 |
+
"grad_norm": 18.75,
|
| 1218 |
+
"learning_rate": 1.5859259259259262e-05,
|
| 1219 |
+
"loss": 16.5571,
|
| 1220 |
+
"mean_token_accuracy": 0.7727442421019077,
|
| 1221 |
+
"num_tokens": 24466132.0,
|
| 1222 |
+
"step": 860
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"entropy": 0.9378596719354391,
|
| 1226 |
+
"epoch": 1.392,
|
| 1227 |
+
"grad_norm": 18.875,
|
| 1228 |
+
"learning_rate": 1.5785185185185185e-05,
|
| 1229 |
+
"loss": 15.5497,
|
| 1230 |
+
"mean_token_accuracy": 0.783870642259717,
|
| 1231 |
+
"num_tokens": 24748043.0,
|
| 1232 |
+
"step": 870
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"entropy": 0.902330132946372,
|
| 1236 |
+
"epoch": 1.408,
|
| 1237 |
+
"grad_norm": 17.0,
|
| 1238 |
+
"learning_rate": 1.571111111111111e-05,
|
| 1239 |
+
"loss": 15.2254,
|
| 1240 |
+
"mean_token_accuracy": 0.7886204585433007,
|
| 1241 |
+
"num_tokens": 25036674.0,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"entropy": 0.9116618972271681,
|
| 1246 |
+
"epoch": 1.424,
|
| 1247 |
+
"grad_norm": 12.125,
|
| 1248 |
+
"learning_rate": 1.5637037037037038e-05,
|
| 1249 |
+
"loss": 15.1439,
|
| 1250 |
+
"mean_token_accuracy": 0.789288330078125,
|
| 1251 |
+
"num_tokens": 25324579.0,
|
| 1252 |
+
"step": 890
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"entropy": 0.9361150417476892,
|
| 1256 |
+
"epoch": 1.44,
|
| 1257 |
+
"grad_norm": 15.3125,
|
| 1258 |
+
"learning_rate": 1.5562962962962965e-05,
|
| 1259 |
+
"loss": 15.6872,
|
| 1260 |
+
"mean_token_accuracy": 0.7842213969677687,
|
| 1261 |
+
"num_tokens": 25606824.0,
|
| 1262 |
+
"step": 900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.44,
|
| 1266 |
+
"eval_biology_entropy": 3.413121500015259,
|
| 1267 |
+
"eval_biology_loss": 3.890507221221924,
|
| 1268 |
+
"eval_biology_mean_token_accuracy": 0.44829230070114134,
|
| 1269 |
+
"eval_biology_num_tokens": 25606824.0,
|
| 1270 |
+
"eval_biology_runtime": 58.6058,
|
| 1271 |
+
"eval_biology_samples_per_second": 8.532,
|
| 1272 |
+
"eval_biology_steps_per_second": 2.133,
|
| 1273 |
+
"step": 900
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 1.44,
|
| 1277 |
+
"eval_chemistry_entropy": 1.990509286403656,
|
| 1278 |
+
"eval_chemistry_loss": 2.1352083683013916,
|
| 1279 |
+
"eval_chemistry_mean_token_accuracy": 0.6389702727794647,
|
| 1280 |
+
"eval_chemistry_num_tokens": 25606824.0,
|
| 1281 |
+
"eval_chemistry_runtime": 72.5787,
|
| 1282 |
+
"eval_chemistry_samples_per_second": 6.889,
|
| 1283 |
+
"eval_chemistry_steps_per_second": 1.722,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.44,
|
| 1288 |
+
"eval_math_entropy": 0.9449980363845826,
|
| 1289 |
+
"eval_math_loss": 0.9944726228713989,
|
| 1290 |
+
"eval_math_mean_token_accuracy": 0.7802386074066162,
|
| 1291 |
+
"eval_math_num_tokens": 25606824.0,
|
| 1292 |
+
"eval_math_runtime": 75.5905,
|
| 1293 |
+
"eval_math_samples_per_second": 6.615,
|
| 1294 |
+
"eval_math_steps_per_second": 1.654,
|
| 1295 |
+
"step": 900
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 1.44,
|
| 1299 |
+
"eval_physics_entropy": 1.601661796092987,
|
| 1300 |
+
"eval_physics_loss": 1.6853703260421753,
|
| 1301 |
+
"eval_physics_mean_token_accuracy": 0.6957120170593262,
|
| 1302 |
+
"eval_physics_num_tokens": 25606824.0,
|
| 1303 |
+
"eval_physics_runtime": 85.8807,
|
| 1304 |
+
"eval_physics_samples_per_second": 5.822,
|
| 1305 |
+
"eval_physics_steps_per_second": 1.456,
|
| 1306 |
+
"step": 900
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"entropy": 0.9202836632728577,
|
| 1310 |
+
"epoch": 1.456,
|
| 1311 |
+
"grad_norm": 13.5625,
|
| 1312 |
+
"learning_rate": 1.548888888888889e-05,
|
| 1313 |
+
"loss": 15.2958,
|
| 1314 |
+
"mean_token_accuracy": 0.7856416959315539,
|
| 1315 |
+
"num_tokens": 25886396.0,
|
| 1316 |
+
"step": 910
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"entropy": 0.9233658915385604,
|
| 1320 |
+
"epoch": 1.472,
|
| 1321 |
+
"grad_norm": 13.25,
|
| 1322 |
+
"learning_rate": 1.5414814814814814e-05,
|
| 1323 |
+
"loss": 15.4872,
|
| 1324 |
+
"mean_token_accuracy": 0.784604623541236,
|
| 1325 |
+
"num_tokens": 26163618.0,
|
| 1326 |
+
"step": 920
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"entropy": 0.9581601906567812,
|
| 1330 |
+
"epoch": 1.488,
|
| 1331 |
+
"grad_norm": 13.8125,
|
| 1332 |
+
"learning_rate": 1.5340740740740744e-05,
|
| 1333 |
+
"loss": 16.0305,
|
| 1334 |
+
"mean_token_accuracy": 0.7800706576555967,
|
| 1335 |
+
"num_tokens": 26438338.0,
|
| 1336 |
+
"step": 930
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"entropy": 0.9096721589565278,
|
| 1340 |
+
"epoch": 1.504,
|
| 1341 |
+
"grad_norm": 17.375,
|
| 1342 |
+
"learning_rate": 1.5266666666666667e-05,
|
| 1343 |
+
"loss": 15.1674,
|
| 1344 |
+
"mean_token_accuracy": 0.7894522257149219,
|
| 1345 |
+
"num_tokens": 26729255.0,
|
| 1346 |
+
"step": 940
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"entropy": 0.9369884114712477,
|
| 1350 |
+
"epoch": 1.52,
|
| 1351 |
+
"grad_norm": 13.5625,
|
| 1352 |
+
"learning_rate": 1.5192592592592594e-05,
|
| 1353 |
+
"loss": 15.7623,
|
| 1354 |
+
"mean_token_accuracy": 0.783218951150775,
|
| 1355 |
+
"num_tokens": 27017935.0,
|
| 1356 |
+
"step": 950
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"entropy": 0.9323047822341323,
|
| 1360 |
+
"epoch": 1.536,
|
| 1361 |
+
"grad_norm": 13.75,
|
| 1362 |
+
"learning_rate": 1.5118518518518519e-05,
|
| 1363 |
+
"loss": 15.4618,
|
| 1364 |
+
"mean_token_accuracy": 0.7854165881872177,
|
| 1365 |
+
"num_tokens": 27306339.0,
|
| 1366 |
+
"step": 960
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"entropy": 0.9303993556648493,
|
| 1370 |
+
"epoch": 1.552,
|
| 1371 |
+
"grad_norm": 16.0,
|
| 1372 |
+
"learning_rate": 1.5044444444444445e-05,
|
| 1373 |
+
"loss": 15.4762,
|
| 1374 |
+
"mean_token_accuracy": 0.7855500593781471,
|
| 1375 |
+
"num_tokens": 27591959.0,
|
| 1376 |
+
"step": 970
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"entropy": 0.8964169017970562,
|
| 1380 |
+
"epoch": 1.568,
|
| 1381 |
+
"grad_norm": 12.9375,
|
| 1382 |
+
"learning_rate": 1.497037037037037e-05,
|
| 1383 |
+
"loss": 15.0334,
|
| 1384 |
+
"mean_token_accuracy": 0.7918794117867947,
|
| 1385 |
+
"num_tokens": 27884398.0,
|
| 1386 |
+
"step": 980
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"entropy": 0.9186629202216864,
|
| 1390 |
+
"epoch": 1.584,
|
| 1391 |
+
"grad_norm": 16.375,
|
| 1392 |
+
"learning_rate": 1.4896296296296298e-05,
|
| 1393 |
+
"loss": 15.3975,
|
| 1394 |
+
"mean_token_accuracy": 0.7869536556303501,
|
| 1395 |
+
"num_tokens": 28171274.0,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"entropy": 0.9073497140780091,
|
| 1400 |
+
"epoch": 1.6,
|
| 1401 |
+
"grad_norm": 18.5,
|
| 1402 |
+
"learning_rate": 1.4822222222222225e-05,
|
| 1403 |
+
"loss": 15.2016,
|
| 1404 |
+
"mean_token_accuracy": 0.7862915009260177,
|
| 1405 |
+
"num_tokens": 28457624.0,
|
| 1406 |
+
"step": 1000
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 1.6,
|
| 1410 |
+
"eval_biology_entropy": 3.321074670791626,
|
| 1411 |
+
"eval_biology_loss": 3.887943983078003,
|
| 1412 |
+
"eval_biology_mean_token_accuracy": 0.44904091477394104,
|
| 1413 |
+
"eval_biology_num_tokens": 28457624.0,
|
| 1414 |
+
"eval_biology_runtime": 58.6475,
|
| 1415 |
+
"eval_biology_samples_per_second": 8.526,
|
| 1416 |
+
"eval_biology_steps_per_second": 2.131,
|
| 1417 |
+
"step": 1000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.6,
|
| 1421 |
+
"eval_chemistry_entropy": 1.93967240858078,
|
| 1422 |
+
"eval_chemistry_loss": 2.1321117877960205,
|
| 1423 |
+
"eval_chemistry_mean_token_accuracy": 0.6390365164279938,
|
| 1424 |
+
"eval_chemistry_num_tokens": 28457624.0,
|
| 1425 |
+
"eval_chemistry_runtime": 71.9495,
|
| 1426 |
+
"eval_chemistry_samples_per_second": 6.949,
|
| 1427 |
+
"eval_chemistry_steps_per_second": 1.737,
|
| 1428 |
+
"step": 1000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 1.6,
|
| 1432 |
+
"eval_math_entropy": 0.9376819987297058,
|
| 1433 |
+
"eval_math_loss": 0.9886224269866943,
|
| 1434 |
+
"eval_math_mean_token_accuracy": 0.7808662033081055,
|
| 1435 |
+
"eval_math_num_tokens": 28457624.0,
|
| 1436 |
+
"eval_math_runtime": 74.8917,
|
| 1437 |
+
"eval_math_samples_per_second": 6.676,
|
| 1438 |
+
"eval_math_steps_per_second": 1.669,
|
| 1439 |
+
"step": 1000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.6,
|
| 1443 |
+
"eval_physics_entropy": 1.566874056339264,
|
| 1444 |
+
"eval_physics_loss": 1.6789716482162476,
|
| 1445 |
+
"eval_physics_mean_token_accuracy": 0.696774085521698,
|
| 1446 |
+
"eval_physics_num_tokens": 28457624.0,
|
| 1447 |
+
"eval_physics_runtime": 85.493,
|
| 1448 |
+
"eval_physics_samples_per_second": 5.848,
|
| 1449 |
+
"eval_physics_steps_per_second": 1.462,
|
| 1450 |
+
"step": 1000
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"entropy": 0.9235946271568537,
|
| 1454 |
+
"epoch": 1.616,
|
| 1455 |
+
"grad_norm": 14.4375,
|
| 1456 |
+
"learning_rate": 1.474814814814815e-05,
|
| 1457 |
+
"loss": 15.3986,
|
| 1458 |
+
"mean_token_accuracy": 0.786095867305994,
|
| 1459 |
+
"num_tokens": 28743099.0,
|
| 1460 |
+
"step": 1010
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"entropy": 0.9303552206605673,
|
| 1464 |
+
"epoch": 1.6320000000000001,
|
| 1465 |
+
"grad_norm": 15.5625,
|
| 1466 |
+
"learning_rate": 1.4674074074074076e-05,
|
| 1467 |
+
"loss": 15.5645,
|
| 1468 |
+
"mean_token_accuracy": 0.7829754035919905,
|
| 1469 |
+
"num_tokens": 29017297.0,
|
| 1470 |
+
"step": 1020
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"entropy": 0.9154788227751851,
|
| 1474 |
+
"epoch": 1.6480000000000001,
|
| 1475 |
+
"grad_norm": 15.4375,
|
| 1476 |
+
"learning_rate": 1.46e-05,
|
| 1477 |
+
"loss": 15.2985,
|
| 1478 |
+
"mean_token_accuracy": 0.7891919452697038,
|
| 1479 |
+
"num_tokens": 29303707.0,
|
| 1480 |
+
"step": 1030
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"entropy": 0.9193036049604416,
|
| 1484 |
+
"epoch": 1.6640000000000001,
|
| 1485 |
+
"grad_norm": 15.875,
|
| 1486 |
+
"learning_rate": 1.4525925925925927e-05,
|
| 1487 |
+
"loss": 15.235,
|
| 1488 |
+
"mean_token_accuracy": 0.7860149636864662,
|
| 1489 |
+
"num_tokens": 29597156.0,
|
| 1490 |
+
"step": 1040
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"entropy": 0.9248684361577034,
|
| 1494 |
+
"epoch": 1.6800000000000002,
|
| 1495 |
+
"grad_norm": 13.5,
|
| 1496 |
+
"learning_rate": 1.4451851851851852e-05,
|
| 1497 |
+
"loss": 15.5595,
|
| 1498 |
+
"mean_token_accuracy": 0.7858113132417202,
|
| 1499 |
+
"num_tokens": 29883879.0,
|
| 1500 |
+
"step": 1050
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"entropy": 0.9327945580706001,
|
| 1504 |
+
"epoch": 1.696,
|
| 1505 |
+
"grad_norm": 14.1875,
|
| 1506 |
+
"learning_rate": 1.4377777777777779e-05,
|
| 1507 |
+
"loss": 15.5428,
|
| 1508 |
+
"mean_token_accuracy": 0.7853838302195072,
|
| 1509 |
+
"num_tokens": 30165760.0,
|
| 1510 |
+
"step": 1060
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"entropy": 0.9061191271990537,
|
| 1514 |
+
"epoch": 1.712,
|
| 1515 |
+
"grad_norm": 15.5625,
|
| 1516 |
+
"learning_rate": 1.4303703703703703e-05,
|
| 1517 |
+
"loss": 15.1161,
|
| 1518 |
+
"mean_token_accuracy": 0.7896213702857494,
|
| 1519 |
+
"num_tokens": 30460367.0,
|
| 1520 |
+
"step": 1070
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"entropy": 0.9054378770291805,
|
| 1524 |
+
"epoch": 1.728,
|
| 1525 |
+
"grad_norm": 19.625,
|
| 1526 |
+
"learning_rate": 1.4229629629629632e-05,
|
| 1527 |
+
"loss": 15.098,
|
| 1528 |
+
"mean_token_accuracy": 0.7892790406942367,
|
| 1529 |
+
"num_tokens": 30739137.0,
|
| 1530 |
+
"step": 1080
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"entropy": 0.9035517876967788,
|
| 1534 |
+
"epoch": 1.744,
|
| 1535 |
+
"grad_norm": 16.25,
|
| 1536 |
+
"learning_rate": 1.4155555555555556e-05,
|
| 1537 |
+
"loss": 15.2616,
|
| 1538 |
+
"mean_token_accuracy": 0.7869241286069155,
|
| 1539 |
+
"num_tokens": 31022663.0,
|
| 1540 |
+
"step": 1090
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"entropy": 0.9091449549421668,
|
| 1544 |
+
"epoch": 1.76,
|
| 1545 |
+
"grad_norm": 13.75,
|
| 1546 |
+
"learning_rate": 1.4081481481481483e-05,
|
| 1547 |
+
"loss": 15.1422,
|
| 1548 |
+
"mean_token_accuracy": 0.7880128789693117,
|
| 1549 |
+
"num_tokens": 31306494.0,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.76,
|
| 1554 |
+
"eval_biology_entropy": 3.356640110015869,
|
| 1555 |
+
"eval_biology_loss": 3.8688619136810303,
|
| 1556 |
+
"eval_biology_mean_token_accuracy": 0.45002903842926023,
|
| 1557 |
+
"eval_biology_num_tokens": 31306494.0,
|
| 1558 |
+
"eval_biology_runtime": 58.9193,
|
| 1559 |
+
"eval_biology_samples_per_second": 8.486,
|
| 1560 |
+
"eval_biology_steps_per_second": 2.122,
|
| 1561 |
+
"step": 1100
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.76,
|
| 1565 |
+
"eval_chemistry_entropy": 1.9536669540405274,
|
| 1566 |
+
"eval_chemistry_loss": 2.1165969371795654,
|
| 1567 |
+
"eval_chemistry_mean_token_accuracy": 0.6407591633796692,
|
| 1568 |
+
"eval_chemistry_num_tokens": 31306494.0,
|
| 1569 |
+
"eval_chemistry_runtime": 71.8391,
|
| 1570 |
+
"eval_chemistry_samples_per_second": 6.96,
|
| 1571 |
+
"eval_chemistry_steps_per_second": 1.74,
|
| 1572 |
+
"step": 1100
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 1.76,
|
| 1576 |
+
"eval_math_entropy": 0.9424327955245971,
|
| 1577 |
+
"eval_math_loss": 0.9843871593475342,
|
| 1578 |
+
"eval_math_mean_token_accuracy": 0.7814709210395813,
|
| 1579 |
+
"eval_math_num_tokens": 31306494.0,
|
| 1580 |
+
"eval_math_runtime": 74.8286,
|
| 1581 |
+
"eval_math_samples_per_second": 6.682,
|
| 1582 |
+
"eval_math_steps_per_second": 1.67,
|
| 1583 |
+
"step": 1100
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 1.76,
|
| 1587 |
+
"eval_physics_entropy": 1.5812338199615479,
|
| 1588 |
+
"eval_physics_loss": 1.6700222492218018,
|
| 1589 |
+
"eval_physics_mean_token_accuracy": 0.697830177783966,
|
| 1590 |
+
"eval_physics_num_tokens": 31306494.0,
|
| 1591 |
+
"eval_physics_runtime": 85.7072,
|
| 1592 |
+
"eval_physics_samples_per_second": 5.834,
|
| 1593 |
+
"eval_physics_steps_per_second": 1.458,
|
| 1594 |
+
"step": 1100
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"entropy": 0.924960282817483,
|
| 1598 |
+
"epoch": 1.776,
|
| 1599 |
+
"grad_norm": 16.5,
|
| 1600 |
+
"learning_rate": 1.400740740740741e-05,
|
| 1601 |
+
"loss": 15.3105,
|
| 1602 |
+
"mean_token_accuracy": 0.7877573404461146,
|
| 1603 |
+
"num_tokens": 31595542.0,
|
| 1604 |
+
"step": 1110
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"entropy": 0.9376434125006199,
|
| 1608 |
+
"epoch": 1.792,
|
| 1609 |
+
"grad_norm": 14.9375,
|
| 1610 |
+
"learning_rate": 1.3933333333333334e-05,
|
| 1611 |
+
"loss": 15.591,
|
| 1612 |
+
"mean_token_accuracy": 0.7839235190302134,
|
| 1613 |
+
"num_tokens": 31881189.0,
|
| 1614 |
+
"step": 1120
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"entropy": 0.9261387910693883,
|
| 1618 |
+
"epoch": 1.808,
|
| 1619 |
+
"grad_norm": 14.5,
|
| 1620 |
+
"learning_rate": 1.385925925925926e-05,
|
| 1621 |
+
"loss": 15.4967,
|
| 1622 |
+
"mean_token_accuracy": 0.7858365088701248,
|
| 1623 |
+
"num_tokens": 32164196.0,
|
| 1624 |
+
"step": 1130
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"entropy": 0.881668720766902,
|
| 1628 |
+
"epoch": 1.8239999999999998,
|
| 1629 |
+
"grad_norm": 13.3125,
|
| 1630 |
+
"learning_rate": 1.3785185185185186e-05,
|
| 1631 |
+
"loss": 14.7294,
|
| 1632 |
+
"mean_token_accuracy": 0.793937573581934,
|
| 1633 |
+
"num_tokens": 32441530.0,
|
| 1634 |
+
"step": 1140
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"entropy": 0.9093685988336802,
|
| 1638 |
+
"epoch": 1.8399999999999999,
|
| 1639 |
+
"grad_norm": 15.1875,
|
| 1640 |
+
"learning_rate": 1.3711111111111112e-05,
|
| 1641 |
+
"loss": 15.1805,
|
| 1642 |
+
"mean_token_accuracy": 0.7902645103633403,
|
| 1643 |
+
"num_tokens": 32723145.0,
|
| 1644 |
+
"step": 1150
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"entropy": 0.9354806829243898,
|
| 1648 |
+
"epoch": 1.8559999999999999,
|
| 1649 |
+
"grad_norm": 16.0,
|
| 1650 |
+
"learning_rate": 1.3637037037037037e-05,
|
| 1651 |
+
"loss": 15.5881,
|
| 1652 |
+
"mean_token_accuracy": 0.7842747800052166,
|
| 1653 |
+
"num_tokens": 33011263.0,
|
| 1654 |
+
"step": 1160
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"entropy": 0.9051053501665592,
|
| 1658 |
+
"epoch": 1.8719999999999999,
|
| 1659 |
+
"grad_norm": 14.1875,
|
| 1660 |
+
"learning_rate": 1.3562962962962965e-05,
|
| 1661 |
+
"loss": 15.1055,
|
| 1662 |
+
"mean_token_accuracy": 0.7910506792366505,
|
| 1663 |
+
"num_tokens": 33298921.0,
|
| 1664 |
+
"step": 1170
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"entropy": 0.9241808012127877,
|
| 1668 |
+
"epoch": 1.888,
|
| 1669 |
+
"grad_norm": 17.25,
|
| 1670 |
+
"learning_rate": 1.3488888888888888e-05,
|
| 1671 |
+
"loss": 15.4978,
|
| 1672 |
+
"mean_token_accuracy": 0.7850798547267914,
|
| 1673 |
+
"num_tokens": 33576243.0,
|
| 1674 |
+
"step": 1180
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"entropy": 0.9191399356350303,
|
| 1678 |
+
"epoch": 1.904,
|
| 1679 |
+
"grad_norm": 13.5,
|
| 1680 |
+
"learning_rate": 1.3414814814814817e-05,
|
| 1681 |
+
"loss": 15.3401,
|
| 1682 |
+
"mean_token_accuracy": 0.7869909830391407,
|
| 1683 |
+
"num_tokens": 33850968.0,
|
| 1684 |
+
"step": 1190
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"entropy": 0.8988351378589868,
|
| 1688 |
+
"epoch": 1.92,
|
| 1689 |
+
"grad_norm": 14.3125,
|
| 1690 |
+
"learning_rate": 1.3340740740740741e-05,
|
| 1691 |
+
"loss": 15.0702,
|
| 1692 |
+
"mean_token_accuracy": 0.7889779832214117,
|
| 1693 |
+
"num_tokens": 34128558.0,
|
| 1694 |
+
"step": 1200
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 1.92,
|
| 1698 |
+
"eval_biology_entropy": 3.342221477508545,
|
| 1699 |
+
"eval_biology_loss": 3.8490023612976074,
|
| 1700 |
+
"eval_biology_mean_token_accuracy": 0.45095433259010315,
|
| 1701 |
+
"eval_biology_num_tokens": 34128558.0,
|
| 1702 |
+
"eval_biology_runtime": 58.5045,
|
| 1703 |
+
"eval_biology_samples_per_second": 8.546,
|
| 1704 |
+
"eval_biology_steps_per_second": 2.137,
|
| 1705 |
+
"step": 1200
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 1.92,
|
| 1709 |
+
"eval_chemistry_entropy": 1.9384882287979126,
|
| 1710 |
+
"eval_chemistry_loss": 2.1126294136047363,
|
| 1711 |
+
"eval_chemistry_mean_token_accuracy": 0.6409776957035065,
|
| 1712 |
+
"eval_chemistry_num_tokens": 34128558.0,
|
| 1713 |
+
"eval_chemistry_runtime": 72.4461,
|
| 1714 |
+
"eval_chemistry_samples_per_second": 6.902,
|
| 1715 |
+
"eval_chemistry_steps_per_second": 1.725,
|
| 1716 |
+
"step": 1200
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 1.92,
|
| 1720 |
+
"eval_math_entropy": 0.9290759258270264,
|
| 1721 |
+
"eval_math_loss": 0.979581892490387,
|
| 1722 |
+
"eval_math_mean_token_accuracy": 0.7822537088394165,
|
| 1723 |
+
"eval_math_num_tokens": 34128558.0,
|
| 1724 |
+
"eval_math_runtime": 75.8783,
|
| 1725 |
+
"eval_math_samples_per_second": 6.589,
|
| 1726 |
+
"eval_math_steps_per_second": 1.647,
|
| 1727 |
+
"step": 1200
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 1.92,
|
| 1731 |
+
"eval_physics_entropy": 1.5694564418792725,
|
| 1732 |
+
"eval_physics_loss": 1.672090768814087,
|
| 1733 |
+
"eval_physics_mean_token_accuracy": 0.6977639694213867,
|
| 1734 |
+
"eval_physics_num_tokens": 34128558.0,
|
| 1735 |
+
"eval_physics_runtime": 85.8572,
|
| 1736 |
+
"eval_physics_samples_per_second": 5.824,
|
| 1737 |
+
"eval_physics_steps_per_second": 1.456,
|
| 1738 |
+
"step": 1200
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"entropy": 0.9215047180652618,
|
| 1742 |
+
"epoch": 1.936,
|
| 1743 |
+
"grad_norm": 18.0,
|
| 1744 |
+
"learning_rate": 1.3266666666666668e-05,
|
| 1745 |
+
"loss": 15.3559,
|
| 1746 |
+
"mean_token_accuracy": 0.7858322024345398,
|
| 1747 |
+
"num_tokens": 34408056.0,
|
| 1748 |
+
"step": 1210
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"entropy": 0.9102119285613298,
|
| 1752 |
+
"epoch": 1.952,
|
| 1753 |
+
"grad_norm": 15.5,
|
| 1754 |
+
"learning_rate": 1.3192592592592594e-05,
|
| 1755 |
+
"loss": 15.183,
|
| 1756 |
+
"mean_token_accuracy": 0.7886962842196226,
|
| 1757 |
+
"num_tokens": 34684679.0,
|
| 1758 |
+
"step": 1220
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"entropy": 0.9319753609597683,
|
| 1762 |
+
"epoch": 1.968,
|
| 1763 |
+
"grad_norm": 16.25,
|
| 1764 |
+
"learning_rate": 1.311851851851852e-05,
|
| 1765 |
+
"loss": 15.6225,
|
| 1766 |
+
"mean_token_accuracy": 0.7829237889498473,
|
| 1767 |
+
"num_tokens": 34971038.0,
|
| 1768 |
+
"step": 1230
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"entropy": 0.9107826724648476,
|
| 1772 |
+
"epoch": 1.984,
|
| 1773 |
+
"grad_norm": 15.3125,
|
| 1774 |
+
"learning_rate": 1.3044444444444446e-05,
|
| 1775 |
+
"loss": 15.0632,
|
| 1776 |
+
"mean_token_accuracy": 0.7896871905773878,
|
| 1777 |
+
"num_tokens": 35262281.0,
|
| 1778 |
+
"step": 1240
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"entropy": 0.8834621507674456,
|
| 1782 |
+
"epoch": 2.0,
|
| 1783 |
+
"grad_norm": 15.8125,
|
| 1784 |
+
"learning_rate": 1.297037037037037e-05,
|
| 1785 |
+
"loss": 14.8785,
|
| 1786 |
+
"mean_token_accuracy": 0.7918516155332327,
|
| 1787 |
+
"num_tokens": 35560864.0,
|
| 1788 |
+
"step": 1250
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"entropy": 0.9305843397974968,
|
| 1792 |
+
"epoch": 2.016,
|
| 1793 |
+
"grad_norm": 13.0,
|
| 1794 |
+
"learning_rate": 1.2896296296296299e-05,
|
| 1795 |
+
"loss": 15.4293,
|
| 1796 |
+
"mean_token_accuracy": 0.786306445300579,
|
| 1797 |
+
"num_tokens": 35846704.0,
|
| 1798 |
+
"step": 1260
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"entropy": 0.9240877356380224,
|
| 1802 |
+
"epoch": 2.032,
|
| 1803 |
+
"grad_norm": 18.375,
|
| 1804 |
+
"learning_rate": 1.2822222222222222e-05,
|
| 1805 |
+
"loss": 15.3222,
|
| 1806 |
+
"mean_token_accuracy": 0.7866721350699664,
|
| 1807 |
+
"num_tokens": 36128775.0,
|
| 1808 |
+
"step": 1270
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"entropy": 0.88644311837852,
|
| 1812 |
+
"epoch": 2.048,
|
| 1813 |
+
"grad_norm": 14.9375,
|
| 1814 |
+
"learning_rate": 1.274814814814815e-05,
|
| 1815 |
+
"loss": 14.7864,
|
| 1816 |
+
"mean_token_accuracy": 0.7933717489242553,
|
| 1817 |
+
"num_tokens": 36419504.0,
|
| 1818 |
+
"step": 1280
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"entropy": 0.8904413890093565,
|
| 1822 |
+
"epoch": 2.064,
|
| 1823 |
+
"grad_norm": 13.8125,
|
| 1824 |
+
"learning_rate": 1.2674074074074075e-05,
|
| 1825 |
+
"loss": 14.9247,
|
| 1826 |
+
"mean_token_accuracy": 0.7927331611514091,
|
| 1827 |
+
"num_tokens": 36706816.0,
|
| 1828 |
+
"step": 1290
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"entropy": 0.9007925825193525,
|
| 1832 |
+
"epoch": 2.08,
|
| 1833 |
+
"grad_norm": 15.25,
|
| 1834 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 1835 |
+
"loss": 14.9352,
|
| 1836 |
+
"mean_token_accuracy": 0.7917296580970288,
|
| 1837 |
+
"num_tokens": 36988475.0,
|
| 1838 |
+
"step": 1300
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 2.08,
|
| 1842 |
+
"eval_biology_entropy": 3.2783792409896853,
|
| 1843 |
+
"eval_biology_loss": 3.845946788787842,
|
| 1844 |
+
"eval_biology_mean_token_accuracy": 0.4511831784248352,
|
| 1845 |
+
"eval_biology_num_tokens": 36988475.0,
|
| 1846 |
+
"eval_biology_runtime": 58.3894,
|
| 1847 |
+
"eval_biology_samples_per_second": 8.563,
|
| 1848 |
+
"eval_biology_steps_per_second": 2.141,
|
| 1849 |
+
"step": 1300
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 2.08,
|
| 1853 |
+
"eval_chemistry_entropy": 1.9022014627456665,
|
| 1854 |
+
"eval_chemistry_loss": 2.1077961921691895,
|
| 1855 |
+
"eval_chemistry_mean_token_accuracy": 0.641357579946518,
|
| 1856 |
+
"eval_chemistry_num_tokens": 36988475.0,
|
| 1857 |
+
"eval_chemistry_runtime": 71.9668,
|
| 1858 |
+
"eval_chemistry_samples_per_second": 6.948,
|
| 1859 |
+
"eval_chemistry_steps_per_second": 1.737,
|
| 1860 |
+
"step": 1300
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 2.08,
|
| 1864 |
+
"eval_math_entropy": 0.9213895635604858,
|
| 1865 |
+
"eval_math_loss": 0.9778892993927002,
|
| 1866 |
+
"eval_math_mean_token_accuracy": 0.7825007963180542,
|
| 1867 |
+
"eval_math_num_tokens": 36988475.0,
|
| 1868 |
+
"eval_math_runtime": 75.2485,
|
| 1869 |
+
"eval_math_samples_per_second": 6.645,
|
| 1870 |
+
"eval_math_steps_per_second": 1.661,
|
| 1871 |
+
"step": 1300
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.08,
|
| 1875 |
+
"eval_physics_entropy": 1.5423394293785095,
|
| 1876 |
+
"eval_physics_loss": 1.6658241748809814,
|
| 1877 |
+
"eval_physics_mean_token_accuracy": 0.6982630844116211,
|
| 1878 |
+
"eval_physics_num_tokens": 36988475.0,
|
| 1879 |
+
"eval_physics_runtime": 85.7778,
|
| 1880 |
+
"eval_physics_samples_per_second": 5.829,
|
| 1881 |
+
"eval_physics_steps_per_second": 1.457,
|
| 1882 |
+
"step": 1300
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"entropy": 0.9216024700552226,
|
| 1886 |
+
"epoch": 2.096,
|
| 1887 |
+
"grad_norm": 15.375,
|
| 1888 |
+
"learning_rate": 1.2525925925925928e-05,
|
| 1889 |
+
"loss": 15.3605,
|
| 1890 |
+
"mean_token_accuracy": 0.7878219708800316,
|
| 1891 |
+
"num_tokens": 37270131.0,
|
| 1892 |
+
"step": 1310
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"entropy": 0.8836073707789183,
|
| 1896 |
+
"epoch": 2.112,
|
| 1897 |
+
"grad_norm": 13.875,
|
| 1898 |
+
"learning_rate": 1.2451851851851853e-05,
|
| 1899 |
+
"loss": 14.7465,
|
| 1900 |
+
"mean_token_accuracy": 0.7925711952149868,
|
| 1901 |
+
"num_tokens": 37563537.0,
|
| 1902 |
+
"step": 1320
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"entropy": 0.8804319033399224,
|
| 1906 |
+
"epoch": 2.128,
|
| 1907 |
+
"grad_norm": 15.5625,
|
| 1908 |
+
"learning_rate": 1.237777777777778e-05,
|
| 1909 |
+
"loss": 14.7508,
|
| 1910 |
+
"mean_token_accuracy": 0.7928811598569154,
|
| 1911 |
+
"num_tokens": 37843959.0,
|
| 1912 |
+
"step": 1330
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"entropy": 0.8922224432229996,
|
| 1916 |
+
"epoch": 2.144,
|
| 1917 |
+
"grad_norm": 16.125,
|
| 1918 |
+
"learning_rate": 1.2303703703703704e-05,
|
| 1919 |
+
"loss": 14.9258,
|
| 1920 |
+
"mean_token_accuracy": 0.7923558741807938,
|
| 1921 |
+
"num_tokens": 38133092.0,
|
| 1922 |
+
"step": 1340
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"entropy": 0.8934669015929103,
|
| 1926 |
+
"epoch": 2.16,
|
| 1927 |
+
"grad_norm": 15.0,
|
| 1928 |
+
"learning_rate": 1.222962962962963e-05,
|
| 1929 |
+
"loss": 14.9419,
|
| 1930 |
+
"mean_token_accuracy": 0.7924261353909969,
|
| 1931 |
+
"num_tokens": 38421229.0,
|
| 1932 |
+
"step": 1350
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"entropy": 0.8715774483978749,
|
| 1936 |
+
"epoch": 2.176,
|
| 1937 |
+
"grad_norm": 17.75,
|
| 1938 |
+
"learning_rate": 1.2155555555555555e-05,
|
| 1939 |
+
"loss": 14.4375,
|
| 1940 |
+
"mean_token_accuracy": 0.7976760245859623,
|
| 1941 |
+
"num_tokens": 38708043.0,
|
| 1942 |
+
"step": 1360
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"entropy": 0.8904842108488082,
|
| 1946 |
+
"epoch": 2.192,
|
| 1947 |
+
"grad_norm": 14.1875,
|
| 1948 |
+
"learning_rate": 1.2081481481481484e-05,
|
| 1949 |
+
"loss": 14.9715,
|
| 1950 |
+
"mean_token_accuracy": 0.7901181749999523,
|
| 1951 |
+
"num_tokens": 38996930.0,
|
| 1952 |
+
"step": 1370
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"entropy": 0.9079100303351879,
|
| 1956 |
+
"epoch": 2.208,
|
| 1957 |
+
"grad_norm": 17.375,
|
| 1958 |
+
"learning_rate": 1.2007407407407408e-05,
|
| 1959 |
+
"loss": 15.1465,
|
| 1960 |
+
"mean_token_accuracy": 0.7897144511342049,
|
| 1961 |
+
"num_tokens": 39279481.0,
|
| 1962 |
+
"step": 1380
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"entropy": 0.9049843695014715,
|
| 1966 |
+
"epoch": 2.224,
|
| 1967 |
+
"grad_norm": 16.875,
|
| 1968 |
+
"learning_rate": 1.1933333333333335e-05,
|
| 1969 |
+
"loss": 15.0122,
|
| 1970 |
+
"mean_token_accuracy": 0.7912030033767223,
|
| 1971 |
+
"num_tokens": 39569030.0,
|
| 1972 |
+
"step": 1390
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"entropy": 0.8973436402156949,
|
| 1976 |
+
"epoch": 2.24,
|
| 1977 |
+
"grad_norm": 14.9375,
|
| 1978 |
+
"learning_rate": 1.185925925925926e-05,
|
| 1979 |
+
"loss": 14.8965,
|
| 1980 |
+
"mean_token_accuracy": 0.7904580168426036,
|
| 1981 |
+
"num_tokens": 39854873.0,
|
| 1982 |
+
"step": 1400
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 2.24,
|
| 1986 |
+
"eval_biology_entropy": 3.335082673072815,
|
| 1987 |
+
"eval_biology_loss": 3.848658800125122,
|
| 1988 |
+
"eval_biology_mean_token_accuracy": 0.4510736925601959,
|
| 1989 |
+
"eval_biology_num_tokens": 39854873.0,
|
| 1990 |
+
"eval_biology_runtime": 58.072,
|
| 1991 |
+
"eval_biology_samples_per_second": 8.61,
|
| 1992 |
+
"eval_biology_steps_per_second": 2.153,
|
| 1993 |
+
"step": 1400
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 2.24,
|
| 1997 |
+
"eval_chemistry_entropy": 1.9277294964790344,
|
| 1998 |
+
"eval_chemistry_loss": 2.105435371398926,
|
| 1999 |
+
"eval_chemistry_mean_token_accuracy": 0.6416595306396484,
|
| 2000 |
+
"eval_chemistry_num_tokens": 39854873.0,
|
| 2001 |
+
"eval_chemistry_runtime": 71.9974,
|
| 2002 |
+
"eval_chemistry_samples_per_second": 6.945,
|
| 2003 |
+
"eval_chemistry_steps_per_second": 1.736,
|
| 2004 |
+
"step": 1400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 2.24,
|
| 2008 |
+
"eval_math_entropy": 0.923270836353302,
|
| 2009 |
+
"eval_math_loss": 0.9751716256141663,
|
| 2010 |
+
"eval_math_mean_token_accuracy": 0.7829953188896179,
|
| 2011 |
+
"eval_math_num_tokens": 39854873.0,
|
| 2012 |
+
"eval_math_runtime": 75.3644,
|
| 2013 |
+
"eval_math_samples_per_second": 6.634,
|
| 2014 |
+
"eval_math_steps_per_second": 1.659,
|
| 2015 |
+
"step": 1400
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 2.24,
|
| 2019 |
+
"eval_physics_entropy": 1.5597347102165222,
|
| 2020 |
+
"eval_physics_loss": 1.6624547243118286,
|
| 2021 |
+
"eval_physics_mean_token_accuracy": 0.6985624618530274,
|
| 2022 |
+
"eval_physics_num_tokens": 39854873.0,
|
| 2023 |
+
"eval_physics_runtime": 85.8375,
|
| 2024 |
+
"eval_physics_samples_per_second": 5.825,
|
| 2025 |
+
"eval_physics_steps_per_second": 1.456,
|
| 2026 |
+
"step": 1400
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"entropy": 0.8929722145199775,
|
| 2030 |
+
"epoch": 2.2560000000000002,
|
| 2031 |
+
"grad_norm": 14.5,
|
| 2032 |
+
"learning_rate": 1.1785185185185186e-05,
|
| 2033 |
+
"loss": 14.8853,
|
| 2034 |
+
"mean_token_accuracy": 0.7903232526034116,
|
| 2035 |
+
"num_tokens": 40141190.0,
|
| 2036 |
+
"step": 1410
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"entropy": 0.9134282995015383,
|
| 2040 |
+
"epoch": 2.2720000000000002,
|
| 2041 |
+
"grad_norm": 17.375,
|
| 2042 |
+
"learning_rate": 1.1711111111111113e-05,
|
| 2043 |
+
"loss": 15.268,
|
| 2044 |
+
"mean_token_accuracy": 0.7882483631372452,
|
| 2045 |
+
"num_tokens": 40415203.0,
|
| 2046 |
+
"step": 1420
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"entropy": 0.893715601041913,
|
| 2050 |
+
"epoch": 2.288,
|
| 2051 |
+
"grad_norm": 15.3125,
|
| 2052 |
+
"learning_rate": 1.1637037037037037e-05,
|
| 2053 |
+
"loss": 15.0269,
|
| 2054 |
+
"mean_token_accuracy": 0.7883617911487818,
|
| 2055 |
+
"num_tokens": 40702393.0,
|
| 2056 |
+
"step": 1430
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"entropy": 0.9125330992043018,
|
| 2060 |
+
"epoch": 2.304,
|
| 2061 |
+
"grad_norm": 16.5,
|
| 2062 |
+
"learning_rate": 1.1562962962962964e-05,
|
| 2063 |
+
"loss": 15.0516,
|
| 2064 |
+
"mean_token_accuracy": 0.7916629381477833,
|
| 2065 |
+
"num_tokens": 40982775.0,
|
| 2066 |
+
"step": 1440
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"entropy": 0.9030418084934354,
|
| 2070 |
+
"epoch": 2.32,
|
| 2071 |
+
"grad_norm": 16.625,
|
| 2072 |
+
"learning_rate": 1.1488888888888889e-05,
|
| 2073 |
+
"loss": 15.0525,
|
| 2074 |
+
"mean_token_accuracy": 0.7892082892358303,
|
| 2075 |
+
"num_tokens": 41263356.0,
|
| 2076 |
+
"step": 1450
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"entropy": 0.9153900254517794,
|
| 2080 |
+
"epoch": 2.336,
|
| 2081 |
+
"grad_norm": 17.0,
|
| 2082 |
+
"learning_rate": 1.1414814814814817e-05,
|
| 2083 |
+
"loss": 15.3005,
|
| 2084 |
+
"mean_token_accuracy": 0.787101947888732,
|
| 2085 |
+
"num_tokens": 41545235.0,
|
| 2086 |
+
"step": 1460
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"entropy": 0.9137625804170966,
|
| 2090 |
+
"epoch": 2.352,
|
| 2091 |
+
"grad_norm": 15.75,
|
| 2092 |
+
"learning_rate": 1.1340740740740742e-05,
|
| 2093 |
+
"loss": 15.1855,
|
| 2094 |
+
"mean_token_accuracy": 0.7869684521108866,
|
| 2095 |
+
"num_tokens": 41833417.0,
|
| 2096 |
+
"step": 1470
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"entropy": 0.8969733513891697,
|
| 2100 |
+
"epoch": 2.368,
|
| 2101 |
+
"grad_norm": 16.875,
|
| 2102 |
+
"learning_rate": 1.1266666666666668e-05,
|
| 2103 |
+
"loss": 15.0008,
|
| 2104 |
+
"mean_token_accuracy": 0.7912497945129872,
|
| 2105 |
+
"num_tokens": 42117030.0,
|
| 2106 |
+
"step": 1480
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"entropy": 0.9257304556667805,
|
| 2110 |
+
"epoch": 2.384,
|
| 2111 |
+
"grad_norm": 16.125,
|
| 2112 |
+
"learning_rate": 1.1192592592592593e-05,
|
| 2113 |
+
"loss": 15.3741,
|
| 2114 |
+
"mean_token_accuracy": 0.7869575574994088,
|
| 2115 |
+
"num_tokens": 42410990.0,
|
| 2116 |
+
"step": 1490
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"entropy": 0.9102526713162661,
|
| 2120 |
+
"epoch": 2.4,
|
| 2121 |
+
"grad_norm": 26.625,
|
| 2122 |
+
"learning_rate": 1.111851851851852e-05,
|
| 2123 |
+
"loss": 15.2415,
|
| 2124 |
+
"mean_token_accuracy": 0.7886898033320904,
|
| 2125 |
+
"num_tokens": 42691890.0,
|
| 2126 |
+
"step": 1500
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 2.4,
|
| 2130 |
+
"eval_biology_entropy": 3.2777514753341674,
|
| 2131 |
+
"eval_biology_loss": 3.851931095123291,
|
| 2132 |
+
"eval_biology_mean_token_accuracy": 0.4516991958618164,
|
| 2133 |
+
"eval_biology_num_tokens": 42691890.0,
|
| 2134 |
+
"eval_biology_runtime": 58.0868,
|
| 2135 |
+
"eval_biology_samples_per_second": 8.608,
|
| 2136 |
+
"eval_biology_steps_per_second": 2.152,
|
| 2137 |
+
"step": 1500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.4,
|
| 2141 |
+
"eval_chemistry_entropy": 1.8954128623008728,
|
| 2142 |
+
"eval_chemistry_loss": 2.109870672225952,
|
| 2143 |
+
"eval_chemistry_mean_token_accuracy": 0.6416130490303039,
|
| 2144 |
+
"eval_chemistry_num_tokens": 42691890.0,
|
| 2145 |
+
"eval_chemistry_runtime": 71.6131,
|
| 2146 |
+
"eval_chemistry_samples_per_second": 6.982,
|
| 2147 |
+
"eval_chemistry_steps_per_second": 1.745,
|
| 2148 |
+
"step": 1500
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 2.4,
|
| 2152 |
+
"eval_math_entropy": 0.9151335005760193,
|
| 2153 |
+
"eval_math_loss": 0.9732398390769958,
|
| 2154 |
+
"eval_math_mean_token_accuracy": 0.78314342212677,
|
| 2155 |
+
"eval_math_num_tokens": 42691890.0,
|
| 2156 |
+
"eval_math_runtime": 75.1319,
|
| 2157 |
+
"eval_math_samples_per_second": 6.655,
|
| 2158 |
+
"eval_math_steps_per_second": 1.664,
|
| 2159 |
+
"step": 1500
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.4,
|
| 2163 |
+
"eval_physics_entropy": 1.5351521530151366,
|
| 2164 |
+
"eval_physics_loss": 1.664096474647522,
|
| 2165 |
+
"eval_physics_mean_token_accuracy": 0.6988328862190246,
|
| 2166 |
+
"eval_physics_num_tokens": 42691890.0,
|
| 2167 |
+
"eval_physics_runtime": 85.6903,
|
| 2168 |
+
"eval_physics_samples_per_second": 5.835,
|
| 2169 |
+
"eval_physics_steps_per_second": 1.459,
|
| 2170 |
+
"step": 1500
|
| 2171 |
+
}
|
| 2172 |
+
],
|
| 2173 |
+
"logging_steps": 10,
|
| 2174 |
+
"max_steps": 3000,
|
| 2175 |
+
"num_input_tokens_seen": 0,
|
| 2176 |
+
"num_train_epochs": 5,
|
| 2177 |
+
"save_steps": 500,
|
| 2178 |
+
"stateful_callbacks": {
|
| 2179 |
+
"TrainerControl": {
|
| 2180 |
+
"args": {
|
| 2181 |
+
"should_epoch_stop": false,
|
| 2182 |
+
"should_evaluate": false,
|
| 2183 |
+
"should_log": false,
|
| 2184 |
+
"should_save": true,
|
| 2185 |
+
"should_training_stop": false
|
| 2186 |
+
},
|
| 2187 |
+
"attributes": {}
|
| 2188 |
+
}
|
| 2189 |
+
},
|
| 2190 |
+
"total_flos": 4.3482460648143555e+18,
|
| 2191 |
+
"train_batch_size": 4,
|
| 2192 |
+
"trial_name": null,
|
| 2193 |
+
"trial_params": null
|
| 2194 |
+
}
|
checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e80dd07ccbee44f48a5c8e42d7765cf62056a2ac77df56409b070bcc46cb2c63
|
| 3 |
+
size 6417
|