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:d554c799aab91849a27413d9673d4ab6f31cc4481d4610294a8116e14624b610
|
| 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:ba06f84a8fc9e28333b560b0c3a05859ff2294d308cb3d8ad53de48389aca4a9
|
| 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 sha256:0866355e984a6c5e7c0bf9b6e0f3dc7a0f358abf86236152923c2bf1929a2dfd
|
| 3 |
+
size 1793214689
|
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": 0.9826594747602939,
|
| 14 |
+
"epoch": 0.016,
|
| 15 |
+
"grad_norm": 156.0,
|
| 16 |
+
"learning_rate": 6.000000000000001e-07,
|
| 17 |
+
"loss": 33.4095,
|
| 18 |
+
"mean_token_accuracy": 0.6464727316051722,
|
| 19 |
+
"num_tokens": 280941.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.0350239120423794,
|
| 24 |
+
"epoch": 0.032,
|
| 25 |
+
"grad_norm": 138.0,
|
| 26 |
+
"learning_rate": 1.2666666666666669e-06,
|
| 27 |
+
"loss": 33.8115,
|
| 28 |
+
"mean_token_accuracy": 0.6422434952110052,
|
| 29 |
+
"num_tokens": 558056.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.0517900373786688,
|
| 34 |
+
"epoch": 0.048,
|
| 35 |
+
"grad_norm": 120.0,
|
| 36 |
+
"learning_rate": 1.9333333333333336e-06,
|
| 37 |
+
"loss": 32.9081,
|
| 38 |
+
"mean_token_accuracy": 0.6461116325110197,
|
| 39 |
+
"num_tokens": 836753.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.1479264814406633,
|
| 44 |
+
"epoch": 0.064,
|
| 45 |
+
"grad_norm": 96.5,
|
| 46 |
+
"learning_rate": 2.6e-06,
|
| 47 |
+
"loss": 31.9094,
|
| 48 |
+
"mean_token_accuracy": 0.6435929080471396,
|
| 49 |
+
"num_tokens": 1126446.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.285454823821783,
|
| 54 |
+
"epoch": 0.08,
|
| 55 |
+
"grad_norm": 75.5,
|
| 56 |
+
"learning_rate": 3.266666666666667e-06,
|
| 57 |
+
"loss": 29.7071,
|
| 58 |
+
"mean_token_accuracy": 0.6587587364017964,
|
| 59 |
+
"num_tokens": 1413596.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 1.388279514759779,
|
| 64 |
+
"epoch": 0.096,
|
| 65 |
+
"grad_norm": 49.0,
|
| 66 |
+
"learning_rate": 3.9333333333333335e-06,
|
| 67 |
+
"loss": 26.9094,
|
| 68 |
+
"mean_token_accuracy": 0.6805908571928739,
|
| 69 |
+
"num_tokens": 1701193.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 1.4036601662635804,
|
| 74 |
+
"epoch": 0.112,
|
| 75 |
+
"grad_norm": 40.5,
|
| 76 |
+
"learning_rate": 4.600000000000001e-06,
|
| 77 |
+
"loss": 24.8593,
|
| 78 |
+
"mean_token_accuracy": 0.6889510739594698,
|
| 79 |
+
"num_tokens": 1979232.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 1.3617051038891077,
|
| 84 |
+
"epoch": 0.128,
|
| 85 |
+
"grad_norm": 31.0,
|
| 86 |
+
"learning_rate": 5.2666666666666665e-06,
|
| 87 |
+
"loss": 23.138,
|
| 88 |
+
"mean_token_accuracy": 0.701369734108448,
|
| 89 |
+
"num_tokens": 2274177.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.3085643179714679,
|
| 94 |
+
"epoch": 0.144,
|
| 95 |
+
"grad_norm": 26.375,
|
| 96 |
+
"learning_rate": 5.933333333333335e-06,
|
| 97 |
+
"loss": 21.2975,
|
| 98 |
+
"mean_token_accuracy": 0.7145243223756552,
|
| 99 |
+
"num_tokens": 2548445.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.2485874235630035,
|
| 104 |
+
"epoch": 0.16,
|
| 105 |
+
"grad_norm": 28.25,
|
| 106 |
+
"learning_rate": 6.600000000000001e-06,
|
| 107 |
+
"loss": 19.7756,
|
| 108 |
+
"mean_token_accuracy": 0.7278878930956125,
|
| 109 |
+
"num_tokens": 2824418.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.16,
|
| 114 |
+
"eval_biology_entropy": 3.488466117858887,
|
| 115 |
+
"eval_biology_loss": 4.677812576293945,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.3829730122089386,
|
| 117 |
+
"eval_biology_num_tokens": 2824418.0,
|
| 118 |
+
"eval_biology_runtime": 50.0839,
|
| 119 |
+
"eval_biology_samples_per_second": 9.983,
|
| 120 |
+
"eval_biology_steps_per_second": 2.496,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.16,
|
| 125 |
+
"eval_chemistry_entropy": 2.3915954236984254,
|
| 126 |
+
"eval_chemistry_loss": 2.5027005672454834,
|
| 127 |
+
"eval_chemistry_mean_token_accuracy": 0.5624839839935303,
|
| 128 |
+
"eval_chemistry_num_tokens": 2824418.0,
|
| 129 |
+
"eval_chemistry_runtime": 62.0455,
|
| 130 |
+
"eval_chemistry_samples_per_second": 8.059,
|
| 131 |
+
"eval_chemistry_steps_per_second": 2.015,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.16,
|
| 136 |
+
"eval_math_entropy": 1.263984902381897,
|
| 137 |
+
"eval_math_loss": 1.2374709844589233,
|
| 138 |
+
"eval_math_mean_token_accuracy": 0.7252511687278748,
|
| 139 |
+
"eval_math_num_tokens": 2824418.0,
|
| 140 |
+
"eval_math_runtime": 63.6691,
|
| 141 |
+
"eval_math_samples_per_second": 7.853,
|
| 142 |
+
"eval_math_steps_per_second": 1.963,
|
| 143 |
+
"step": 100
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.16,
|
| 147 |
+
"eval_physics_entropy": 1.9593838052749635,
|
| 148 |
+
"eval_physics_loss": 1.9986088275909424,
|
| 149 |
+
"eval_physics_mean_token_accuracy": 0.6278775181770325,
|
| 150 |
+
"eval_physics_num_tokens": 2824418.0,
|
| 151 |
+
"eval_physics_runtime": 72.5186,
|
| 152 |
+
"eval_physics_samples_per_second": 6.895,
|
| 153 |
+
"eval_physics_steps_per_second": 1.724,
|
| 154 |
+
"step": 100
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"entropy": 1.2179695062339306,
|
| 158 |
+
"epoch": 0.176,
|
| 159 |
+
"grad_norm": 22.75,
|
| 160 |
+
"learning_rate": 7.266666666666668e-06,
|
| 161 |
+
"loss": 19.1845,
|
| 162 |
+
"mean_token_accuracy": 0.7315629377961159,
|
| 163 |
+
"num_tokens": 3110313.0,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"entropy": 1.2148031558841468,
|
| 168 |
+
"epoch": 0.192,
|
| 169 |
+
"grad_norm": 21.625,
|
| 170 |
+
"learning_rate": 7.933333333333334e-06,
|
| 171 |
+
"loss": 18.9916,
|
| 172 |
+
"mean_token_accuracy": 0.7303893141448498,
|
| 173 |
+
"num_tokens": 3394170.0,
|
| 174 |
+
"step": 120
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"entropy": 1.1610155675560236,
|
| 178 |
+
"epoch": 0.208,
|
| 179 |
+
"grad_norm": 19.25,
|
| 180 |
+
"learning_rate": 8.6e-06,
|
| 181 |
+
"loss": 18.5031,
|
| 182 |
+
"mean_token_accuracy": 0.7372519567608833,
|
| 183 |
+
"num_tokens": 3673600.0,
|
| 184 |
+
"step": 130
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"entropy": 1.1351299952715634,
|
| 188 |
+
"epoch": 0.224,
|
| 189 |
+
"grad_norm": 21.75,
|
| 190 |
+
"learning_rate": 9.266666666666667e-06,
|
| 191 |
+
"loss": 18.0553,
|
| 192 |
+
"mean_token_accuracy": 0.7397045020014048,
|
| 193 |
+
"num_tokens": 3953732.0,
|
| 194 |
+
"step": 140
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"entropy": 1.0942034516483545,
|
| 198 |
+
"epoch": 0.24,
|
| 199 |
+
"grad_norm": 20.75,
|
| 200 |
+
"learning_rate": 9.933333333333334e-06,
|
| 201 |
+
"loss": 17.3542,
|
| 202 |
+
"mean_token_accuracy": 0.7487591315060854,
|
| 203 |
+
"num_tokens": 4243655.0,
|
| 204 |
+
"step": 150
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"entropy": 1.0573217056691646,
|
| 208 |
+
"epoch": 0.256,
|
| 209 |
+
"grad_norm": 19.75,
|
| 210 |
+
"learning_rate": 1.0600000000000002e-05,
|
| 211 |
+
"loss": 16.8031,
|
| 212 |
+
"mean_token_accuracy": 0.7541655786335468,
|
| 213 |
+
"num_tokens": 4531471.0,
|
| 214 |
+
"step": 160
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"entropy": 1.0622627913951874,
|
| 218 |
+
"epoch": 0.272,
|
| 219 |
+
"grad_norm": 17.25,
|
| 220 |
+
"learning_rate": 1.1266666666666668e-05,
|
| 221 |
+
"loss": 16.9748,
|
| 222 |
+
"mean_token_accuracy": 0.7531674332916737,
|
| 223 |
+
"num_tokens": 4810284.0,
|
| 224 |
+
"step": 170
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"entropy": 1.0466016590595246,
|
| 228 |
+
"epoch": 0.288,
|
| 229 |
+
"grad_norm": 15.8125,
|
| 230 |
+
"learning_rate": 1.1933333333333335e-05,
|
| 231 |
+
"loss": 16.5926,
|
| 232 |
+
"mean_token_accuracy": 0.7563348900526762,
|
| 233 |
+
"num_tokens": 5095104.0,
|
| 234 |
+
"step": 180
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"entropy": 1.0420017059892417,
|
| 238 |
+
"epoch": 0.304,
|
| 239 |
+
"grad_norm": 15.625,
|
| 240 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 241 |
+
"loss": 16.6259,
|
| 242 |
+
"mean_token_accuracy": 0.7554979339241982,
|
| 243 |
+
"num_tokens": 5383732.0,
|
| 244 |
+
"step": 190
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"entropy": 1.0120286911725997,
|
| 248 |
+
"epoch": 0.32,
|
| 249 |
+
"grad_norm": 18.25,
|
| 250 |
+
"learning_rate": 1.3266666666666668e-05,
|
| 251 |
+
"loss": 16.1236,
|
| 252 |
+
"mean_token_accuracy": 0.7615335866808891,
|
| 253 |
+
"num_tokens": 5676334.0,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.32,
|
| 258 |
+
"eval_biology_entropy": 3.398402997970581,
|
| 259 |
+
"eval_biology_loss": 4.174577236175537,
|
| 260 |
+
"eval_biology_mean_token_accuracy": 0.40706401348114013,
|
| 261 |
+
"eval_biology_num_tokens": 5676334.0,
|
| 262 |
+
"eval_biology_runtime": 50.0834,
|
| 263 |
+
"eval_biology_samples_per_second": 9.983,
|
| 264 |
+
"eval_biology_steps_per_second": 2.496,
|
| 265 |
+
"step": 200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.32,
|
| 269 |
+
"eval_chemistry_entropy": 2.1380633115768433,
|
| 270 |
+
"eval_chemistry_loss": 2.1777119636535645,
|
| 271 |
+
"eval_chemistry_mean_token_accuracy": 0.5923420660495758,
|
| 272 |
+
"eval_chemistry_num_tokens": 5676334.0,
|
| 273 |
+
"eval_chemistry_runtime": 62.077,
|
| 274 |
+
"eval_chemistry_samples_per_second": 8.055,
|
| 275 |
+
"eval_chemistry_steps_per_second": 2.014,
|
| 276 |
+
"step": 200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.32,
|
| 280 |
+
"eval_math_entropy": 1.0378505063056946,
|
| 281 |
+
"eval_math_loss": 1.031339168548584,
|
| 282 |
+
"eval_math_mean_token_accuracy": 0.7560033488273621,
|
| 283 |
+
"eval_math_num_tokens": 5676334.0,
|
| 284 |
+
"eval_math_runtime": 63.6444,
|
| 285 |
+
"eval_math_samples_per_second": 7.856,
|
| 286 |
+
"eval_math_steps_per_second": 1.964,
|
| 287 |
+
"step": 200
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.32,
|
| 291 |
+
"eval_physics_entropy": 1.6895962080955504,
|
| 292 |
+
"eval_physics_loss": 1.734020471572876,
|
| 293 |
+
"eval_physics_mean_token_accuracy": 0.6560393133163452,
|
| 294 |
+
"eval_physics_num_tokens": 5676334.0,
|
| 295 |
+
"eval_physics_runtime": 72.4962,
|
| 296 |
+
"eval_physics_samples_per_second": 6.897,
|
| 297 |
+
"eval_physics_steps_per_second": 1.724,
|
| 298 |
+
"step": 200
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"entropy": 1.0268741879612207,
|
| 302 |
+
"epoch": 0.336,
|
| 303 |
+
"grad_norm": 18.0,
|
| 304 |
+
"learning_rate": 1.3933333333333334e-05,
|
| 305 |
+
"loss": 16.46,
|
| 306 |
+
"mean_token_accuracy": 0.7554850295186043,
|
| 307 |
+
"num_tokens": 5958480.0,
|
| 308 |
+
"step": 210
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"entropy": 0.9965500451624394,
|
| 312 |
+
"epoch": 0.352,
|
| 313 |
+
"grad_norm": 16.375,
|
| 314 |
+
"learning_rate": 1.46e-05,
|
| 315 |
+
"loss": 15.7948,
|
| 316 |
+
"mean_token_accuracy": 0.7630108542740345,
|
| 317 |
+
"num_tokens": 6242161.0,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"entropy": 0.994103953614831,
|
| 322 |
+
"epoch": 0.368,
|
| 323 |
+
"grad_norm": 17.875,
|
| 324 |
+
"learning_rate": 1.5266666666666667e-05,
|
| 325 |
+
"loss": 15.8963,
|
| 326 |
+
"mean_token_accuracy": 0.7624320674687624,
|
| 327 |
+
"num_tokens": 6523679.0,
|
| 328 |
+
"step": 230
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"entropy": 0.9746689781546592,
|
| 332 |
+
"epoch": 0.384,
|
| 333 |
+
"grad_norm": 16.875,
|
| 334 |
+
"learning_rate": 1.5933333333333336e-05,
|
| 335 |
+
"loss": 15.5515,
|
| 336 |
+
"mean_token_accuracy": 0.7664125367999077,
|
| 337 |
+
"num_tokens": 6810978.0,
|
| 338 |
+
"step": 240
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"entropy": 1.0081916853785515,
|
| 342 |
+
"epoch": 0.4,
|
| 343 |
+
"grad_norm": 16.5,
|
| 344 |
+
"learning_rate": 1.66e-05,
|
| 345 |
+
"loss": 16.4028,
|
| 346 |
+
"mean_token_accuracy": 0.7586446344852448,
|
| 347 |
+
"num_tokens": 7096903.0,
|
| 348 |
+
"step": 250
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"entropy": 0.9798103597015142,
|
| 352 |
+
"epoch": 0.416,
|
| 353 |
+
"grad_norm": 15.75,
|
| 354 |
+
"learning_rate": 1.726666666666667e-05,
|
| 355 |
+
"loss": 15.6138,
|
| 356 |
+
"mean_token_accuracy": 0.7639607947319746,
|
| 357 |
+
"num_tokens": 7377181.0,
|
| 358 |
+
"step": 260
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"entropy": 0.9465980283915997,
|
| 362 |
+
"epoch": 0.432,
|
| 363 |
+
"grad_norm": 17.625,
|
| 364 |
+
"learning_rate": 1.7933333333333333e-05,
|
| 365 |
+
"loss": 15.2106,
|
| 366 |
+
"mean_token_accuracy": 0.7715261545032263,
|
| 367 |
+
"num_tokens": 7650523.0,
|
| 368 |
+
"step": 270
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"entropy": 0.9358791437000036,
|
| 372 |
+
"epoch": 0.448,
|
| 373 |
+
"grad_norm": 18.875,
|
| 374 |
+
"learning_rate": 1.86e-05,
|
| 375 |
+
"loss": 15.0287,
|
| 376 |
+
"mean_token_accuracy": 0.771769804134965,
|
| 377 |
+
"num_tokens": 7936788.0,
|
| 378 |
+
"step": 280
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"entropy": 0.9421842556446791,
|
| 382 |
+
"epoch": 0.464,
|
| 383 |
+
"grad_norm": 20.0,
|
| 384 |
+
"learning_rate": 1.926666666666667e-05,
|
| 385 |
+
"loss": 15.2138,
|
| 386 |
+
"mean_token_accuracy": 0.7710198521614074,
|
| 387 |
+
"num_tokens": 8223147.0,
|
| 388 |
+
"step": 290
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"entropy": 0.962396053224802,
|
| 392 |
+
"epoch": 0.48,
|
| 393 |
+
"grad_norm": 17.75,
|
| 394 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 395 |
+
"loss": 15.3417,
|
| 396 |
+
"mean_token_accuracy": 0.7693098299205303,
|
| 397 |
+
"num_tokens": 8506030.0,
|
| 398 |
+
"step": 300
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.48,
|
| 402 |
+
"eval_biology_entropy": 3.405539764404297,
|
| 403 |
+
"eval_biology_loss": 3.895674228668213,
|
| 404 |
+
"eval_biology_mean_token_accuracy": 0.4204931092262268,
|
| 405 |
+
"eval_biology_num_tokens": 8506030.0,
|
| 406 |
+
"eval_biology_runtime": 50.1367,
|
| 407 |
+
"eval_biology_samples_per_second": 9.973,
|
| 408 |
+
"eval_biology_steps_per_second": 2.493,
|
| 409 |
+
"step": 300
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.48,
|
| 413 |
+
"eval_chemistry_entropy": 2.0477921390533447,
|
| 414 |
+
"eval_chemistry_loss": 2.050157308578491,
|
| 415 |
+
"eval_chemistry_mean_token_accuracy": 0.6057685849666595,
|
| 416 |
+
"eval_chemistry_num_tokens": 8506030.0,
|
| 417 |
+
"eval_chemistry_runtime": 62.1697,
|
| 418 |
+
"eval_chemistry_samples_per_second": 8.043,
|
| 419 |
+
"eval_chemistry_steps_per_second": 2.011,
|
| 420 |
+
"step": 300
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.48,
|
| 424 |
+
"eval_math_entropy": 0.9374664115905762,
|
| 425 |
+
"eval_math_loss": 0.9544177651405334,
|
| 426 |
+
"eval_math_mean_token_accuracy": 0.7683046078681945,
|
| 427 |
+
"eval_math_num_tokens": 8506030.0,
|
| 428 |
+
"eval_math_runtime": 63.7112,
|
| 429 |
+
"eval_math_samples_per_second": 7.848,
|
| 430 |
+
"eval_math_steps_per_second": 1.962,
|
| 431 |
+
"step": 300
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.48,
|
| 435 |
+
"eval_physics_entropy": 1.6079197850227356,
|
| 436 |
+
"eval_physics_loss": 1.6351362466812134,
|
| 437 |
+
"eval_physics_mean_token_accuracy": 0.6683077902793885,
|
| 438 |
+
"eval_physics_num_tokens": 8506030.0,
|
| 439 |
+
"eval_physics_runtime": 72.5398,
|
| 440 |
+
"eval_physics_samples_per_second": 6.893,
|
| 441 |
+
"eval_physics_steps_per_second": 1.723,
|
| 442 |
+
"step": 300
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"entropy": 0.9215071354061365,
|
| 446 |
+
"epoch": 0.496,
|
| 447 |
+
"grad_norm": 26.875,
|
| 448 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 449 |
+
"loss": 14.792,
|
| 450 |
+
"mean_token_accuracy": 0.7745936445891857,
|
| 451 |
+
"num_tokens": 8788726.0,
|
| 452 |
+
"step": 310
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"entropy": 0.9270129840821028,
|
| 456 |
+
"epoch": 0.512,
|
| 457 |
+
"grad_norm": 17.5,
|
| 458 |
+
"learning_rate": 1.985925925925926e-05,
|
| 459 |
+
"loss": 14.8253,
|
| 460 |
+
"mean_token_accuracy": 0.7746361777186394,
|
| 461 |
+
"num_tokens": 9078039.0,
|
| 462 |
+
"step": 320
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"entropy": 0.899552209302783,
|
| 466 |
+
"epoch": 0.528,
|
| 467 |
+
"grad_norm": 20.125,
|
| 468 |
+
"learning_rate": 1.9785185185185187e-05,
|
| 469 |
+
"loss": 14.5531,
|
| 470 |
+
"mean_token_accuracy": 0.7795046135783196,
|
| 471 |
+
"num_tokens": 9373860.0,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"entropy": 0.9075620226562023,
|
| 476 |
+
"epoch": 0.544,
|
| 477 |
+
"grad_norm": 17.875,
|
| 478 |
+
"learning_rate": 1.971111111111111e-05,
|
| 479 |
+
"loss": 14.713,
|
| 480 |
+
"mean_token_accuracy": 0.777441994100809,
|
| 481 |
+
"num_tokens": 9660940.0,
|
| 482 |
+
"step": 340
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"entropy": 0.9506530221551657,
|
| 486 |
+
"epoch": 0.56,
|
| 487 |
+
"grad_norm": 25.25,
|
| 488 |
+
"learning_rate": 1.963703703703704e-05,
|
| 489 |
+
"loss": 15.1709,
|
| 490 |
+
"mean_token_accuracy": 0.7681664042174816,
|
| 491 |
+
"num_tokens": 9932302.0,
|
| 492 |
+
"step": 350
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"entropy": 0.9100017622113228,
|
| 496 |
+
"epoch": 0.576,
|
| 497 |
+
"grad_norm": 18.0,
|
| 498 |
+
"learning_rate": 1.9562962962962964e-05,
|
| 499 |
+
"loss": 14.6505,
|
| 500 |
+
"mean_token_accuracy": 0.7754532121121883,
|
| 501 |
+
"num_tokens": 10215462.0,
|
| 502 |
+
"step": 360
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"entropy": 0.896766658872366,
|
| 506 |
+
"epoch": 0.592,
|
| 507 |
+
"grad_norm": 16.125,
|
| 508 |
+
"learning_rate": 1.948888888888889e-05,
|
| 509 |
+
"loss": 14.5203,
|
| 510 |
+
"mean_token_accuracy": 0.7769503209739923,
|
| 511 |
+
"num_tokens": 10504396.0,
|
| 512 |
+
"step": 370
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"entropy": 0.9094142084941268,
|
| 516 |
+
"epoch": 0.608,
|
| 517 |
+
"grad_norm": 16.25,
|
| 518 |
+
"learning_rate": 1.9414814814814817e-05,
|
| 519 |
+
"loss": 14.6279,
|
| 520 |
+
"mean_token_accuracy": 0.7770587045699358,
|
| 521 |
+
"num_tokens": 10793126.0,
|
| 522 |
+
"step": 380
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"entropy": 0.8876220740377903,
|
| 526 |
+
"epoch": 0.624,
|
| 527 |
+
"grad_norm": 18.0,
|
| 528 |
+
"learning_rate": 1.9340740740740743e-05,
|
| 529 |
+
"loss": 14.3233,
|
| 530 |
+
"mean_token_accuracy": 0.7783829730004073,
|
| 531 |
+
"num_tokens": 11081768.0,
|
| 532 |
+
"step": 390
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"entropy": 0.88020947445184,
|
| 536 |
+
"epoch": 0.64,
|
| 537 |
+
"grad_norm": 16.75,
|
| 538 |
+
"learning_rate": 1.926666666666667e-05,
|
| 539 |
+
"loss": 14.0524,
|
| 540 |
+
"mean_token_accuracy": 0.780827208608389,
|
| 541 |
+
"num_tokens": 11370320.0,
|
| 542 |
+
"step": 400
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.64,
|
| 546 |
+
"eval_biology_entropy": 3.335602487564087,
|
| 547 |
+
"eval_biology_loss": 3.8722915649414062,
|
| 548 |
+
"eval_biology_mean_token_accuracy": 0.42685190558433533,
|
| 549 |
+
"eval_biology_num_tokens": 11370320.0,
|
| 550 |
+
"eval_biology_runtime": 50.0979,
|
| 551 |
+
"eval_biology_samples_per_second": 9.98,
|
| 552 |
+
"eval_biology_steps_per_second": 2.495,
|
| 553 |
+
"step": 400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.64,
|
| 557 |
+
"eval_chemistry_entropy": 1.9703331971168518,
|
| 558 |
+
"eval_chemistry_loss": 1.9906623363494873,
|
| 559 |
+
"eval_chemistry_mean_token_accuracy": 0.6160369579792022,
|
| 560 |
+
"eval_chemistry_num_tokens": 11370320.0,
|
| 561 |
+
"eval_chemistry_runtime": 62.0625,
|
| 562 |
+
"eval_chemistry_samples_per_second": 8.056,
|
| 563 |
+
"eval_chemistry_steps_per_second": 2.014,
|
| 564 |
+
"step": 400
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.64,
|
| 568 |
+
"eval_math_entropy": 0.9068713026046753,
|
| 569 |
+
"eval_math_loss": 0.910418689250946,
|
| 570 |
+
"eval_math_mean_token_accuracy": 0.7759209518432617,
|
| 571 |
+
"eval_math_num_tokens": 11370320.0,
|
| 572 |
+
"eval_math_runtime": 63.6487,
|
| 573 |
+
"eval_math_samples_per_second": 7.856,
|
| 574 |
+
"eval_math_steps_per_second": 1.964,
|
| 575 |
+
"step": 400
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.64,
|
| 579 |
+
"eval_physics_entropy": 1.5451811146736145,
|
| 580 |
+
"eval_physics_loss": 1.5735290050506592,
|
| 581 |
+
"eval_physics_mean_token_accuracy": 0.6776216115951538,
|
| 582 |
+
"eval_physics_num_tokens": 11370320.0,
|
| 583 |
+
"eval_physics_runtime": 72.4949,
|
| 584 |
+
"eval_physics_samples_per_second": 6.897,
|
| 585 |
+
"eval_physics_steps_per_second": 1.724,
|
| 586 |
+
"step": 400
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"entropy": 0.860218596458435,
|
| 590 |
+
"epoch": 0.656,
|
| 591 |
+
"grad_norm": 15.0625,
|
| 592 |
+
"learning_rate": 1.9192592592592593e-05,
|
| 593 |
+
"loss": 13.8789,
|
| 594 |
+
"mean_token_accuracy": 0.7833294775336981,
|
| 595 |
+
"num_tokens": 11657835.0,
|
| 596 |
+
"step": 410
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"entropy": 0.8948563069105149,
|
| 600 |
+
"epoch": 0.672,
|
| 601 |
+
"grad_norm": 16.5,
|
| 602 |
+
"learning_rate": 1.911851851851852e-05,
|
| 603 |
+
"loss": 14.3792,
|
| 604 |
+
"mean_token_accuracy": 0.7774154812097549,
|
| 605 |
+
"num_tokens": 11949262.0,
|
| 606 |
+
"step": 420
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"entropy": 0.8625882860273123,
|
| 610 |
+
"epoch": 0.688,
|
| 611 |
+
"grad_norm": 14.0,
|
| 612 |
+
"learning_rate": 1.9044444444444446e-05,
|
| 613 |
+
"loss": 13.8761,
|
| 614 |
+
"mean_token_accuracy": 0.7837949499487877,
|
| 615 |
+
"num_tokens": 12227640.0,
|
| 616 |
+
"step": 430
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"entropy": 0.8663179978728295,
|
| 620 |
+
"epoch": 0.704,
|
| 621 |
+
"grad_norm": 16.25,
|
| 622 |
+
"learning_rate": 1.8970370370370372e-05,
|
| 623 |
+
"loss": 13.9767,
|
| 624 |
+
"mean_token_accuracy": 0.7831656277179718,
|
| 625 |
+
"num_tokens": 12516641.0,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"entropy": 0.8871479570865631,
|
| 630 |
+
"epoch": 0.72,
|
| 631 |
+
"grad_norm": 18.125,
|
| 632 |
+
"learning_rate": 1.8896296296296295e-05,
|
| 633 |
+
"loss": 14.3515,
|
| 634 |
+
"mean_token_accuracy": 0.7785210806876421,
|
| 635 |
+
"num_tokens": 12793343.0,
|
| 636 |
+
"step": 450
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"entropy": 0.872298126295209,
|
| 640 |
+
"epoch": 0.736,
|
| 641 |
+
"grad_norm": 15.875,
|
| 642 |
+
"learning_rate": 1.8822222222222225e-05,
|
| 643 |
+
"loss": 14.0105,
|
| 644 |
+
"mean_token_accuracy": 0.7827250022441149,
|
| 645 |
+
"num_tokens": 13077981.0,
|
| 646 |
+
"step": 460
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"entropy": 0.8562807805836201,
|
| 650 |
+
"epoch": 0.752,
|
| 651 |
+
"grad_norm": 17.375,
|
| 652 |
+
"learning_rate": 1.874814814814815e-05,
|
| 653 |
+
"loss": 13.6996,
|
| 654 |
+
"mean_token_accuracy": 0.7862995523959398,
|
| 655 |
+
"num_tokens": 13358957.0,
|
| 656 |
+
"step": 470
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"entropy": 0.8273469803854823,
|
| 660 |
+
"epoch": 0.768,
|
| 661 |
+
"grad_norm": 14.1875,
|
| 662 |
+
"learning_rate": 1.8674074074074075e-05,
|
| 663 |
+
"loss": 13.3777,
|
| 664 |
+
"mean_token_accuracy": 0.7925803154706955,
|
| 665 |
+
"num_tokens": 13653412.0,
|
| 666 |
+
"step": 480
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"entropy": 0.8882941573858261,
|
| 670 |
+
"epoch": 0.784,
|
| 671 |
+
"grad_norm": 17.25,
|
| 672 |
+
"learning_rate": 1.86e-05,
|
| 673 |
+
"loss": 14.3299,
|
| 674 |
+
"mean_token_accuracy": 0.7802201680839062,
|
| 675 |
+
"num_tokens": 13940856.0,
|
| 676 |
+
"step": 490
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"entropy": 0.8893520476296544,
|
| 680 |
+
"epoch": 0.8,
|
| 681 |
+
"grad_norm": 17.25,
|
| 682 |
+
"learning_rate": 1.8525925925925928e-05,
|
| 683 |
+
"loss": 14.3909,
|
| 684 |
+
"mean_token_accuracy": 0.7795332457870245,
|
| 685 |
+
"num_tokens": 14230754.0,
|
| 686 |
+
"step": 500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.8,
|
| 690 |
+
"eval_biology_entropy": 3.442336763381958,
|
| 691 |
+
"eval_biology_loss": 3.7037065029144287,
|
| 692 |
+
"eval_biology_mean_token_accuracy": 0.43281020307540896,
|
| 693 |
+
"eval_biology_num_tokens": 14230754.0,
|
| 694 |
+
"eval_biology_runtime": 49.9381,
|
| 695 |
+
"eval_biology_samples_per_second": 10.012,
|
| 696 |
+
"eval_biology_steps_per_second": 2.503,
|
| 697 |
+
"step": 500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.8,
|
| 701 |
+
"eval_chemistry_entropy": 1.960342116355896,
|
| 702 |
+
"eval_chemistry_loss": 1.9245615005493164,
|
| 703 |
+
"eval_chemistry_mean_token_accuracy": 0.6227234079837799,
|
| 704 |
+
"eval_chemistry_num_tokens": 14230754.0,
|
| 705 |
+
"eval_chemistry_runtime": 61.6487,
|
| 706 |
+
"eval_chemistry_samples_per_second": 8.11,
|
| 707 |
+
"eval_chemistry_steps_per_second": 2.028,
|
| 708 |
+
"step": 500
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.8,
|
| 712 |
+
"eval_math_entropy": 0.8902029480934143,
|
| 713 |
+
"eval_math_loss": 0.885689914226532,
|
| 714 |
+
"eval_math_mean_token_accuracy": 0.7805563545227051,
|
| 715 |
+
"eval_math_num_tokens": 14230754.0,
|
| 716 |
+
"eval_math_runtime": 63.5447,
|
| 717 |
+
"eval_math_samples_per_second": 7.868,
|
| 718 |
+
"eval_math_steps_per_second": 1.967,
|
| 719 |
+
"step": 500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.8,
|
| 723 |
+
"eval_physics_entropy": 1.5454372839927673,
|
| 724 |
+
"eval_physics_loss": 1.5333291292190552,
|
| 725 |
+
"eval_physics_mean_token_accuracy": 0.683074339389801,
|
| 726 |
+
"eval_physics_num_tokens": 14230754.0,
|
| 727 |
+
"eval_physics_runtime": 72.37,
|
| 728 |
+
"eval_physics_samples_per_second": 6.909,
|
| 729 |
+
"eval_physics_steps_per_second": 1.727,
|
| 730 |
+
"step": 500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"entropy": 0.8682117871940136,
|
| 734 |
+
"epoch": 0.816,
|
| 735 |
+
"grad_norm": 16.0,
|
| 736 |
+
"learning_rate": 1.8451851851851855e-05,
|
| 737 |
+
"loss": 13.814,
|
| 738 |
+
"mean_token_accuracy": 0.7841372516006231,
|
| 739 |
+
"num_tokens": 14519893.0,
|
| 740 |
+
"step": 510
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"entropy": 0.8578620586544276,
|
| 744 |
+
"epoch": 0.832,
|
| 745 |
+
"grad_norm": 16.375,
|
| 746 |
+
"learning_rate": 1.8377777777777778e-05,
|
| 747 |
+
"loss": 13.8317,
|
| 748 |
+
"mean_token_accuracy": 0.7859897501766682,
|
| 749 |
+
"num_tokens": 14805088.0,
|
| 750 |
+
"step": 520
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"entropy": 0.8855314686894417,
|
| 754 |
+
"epoch": 0.848,
|
| 755 |
+
"grad_norm": 17.125,
|
| 756 |
+
"learning_rate": 1.8303703703703704e-05,
|
| 757 |
+
"loss": 14.274,
|
| 758 |
+
"mean_token_accuracy": 0.7796536423265934,
|
| 759 |
+
"num_tokens": 15086306.0,
|
| 760 |
+
"step": 530
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"entropy": 0.8731808762997388,
|
| 764 |
+
"epoch": 0.864,
|
| 765 |
+
"grad_norm": 14.1875,
|
| 766 |
+
"learning_rate": 1.822962962962963e-05,
|
| 767 |
+
"loss": 14.0156,
|
| 768 |
+
"mean_token_accuracy": 0.7828668527305126,
|
| 769 |
+
"num_tokens": 15370985.0,
|
| 770 |
+
"step": 540
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"entropy": 0.880347841233015,
|
| 774 |
+
"epoch": 0.88,
|
| 775 |
+
"grad_norm": 19.375,
|
| 776 |
+
"learning_rate": 1.8155555555555557e-05,
|
| 777 |
+
"loss": 14.1502,
|
| 778 |
+
"mean_token_accuracy": 0.7809113275259734,
|
| 779 |
+
"num_tokens": 15653836.0,
|
| 780 |
+
"step": 550
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"entropy": 0.8649558831006289,
|
| 784 |
+
"epoch": 0.896,
|
| 785 |
+
"grad_norm": 20.0,
|
| 786 |
+
"learning_rate": 1.8081481481481484e-05,
|
| 787 |
+
"loss": 13.9752,
|
| 788 |
+
"mean_token_accuracy": 0.7830663651227951,
|
| 789 |
+
"num_tokens": 15932179.0,
|
| 790 |
+
"step": 560
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"entropy": 0.8504776157438755,
|
| 794 |
+
"epoch": 0.912,
|
| 795 |
+
"grad_norm": 16.625,
|
| 796 |
+
"learning_rate": 1.800740740740741e-05,
|
| 797 |
+
"loss": 13.5728,
|
| 798 |
+
"mean_token_accuracy": 0.7864992685616017,
|
| 799 |
+
"num_tokens": 16219842.0,
|
| 800 |
+
"step": 570
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"entropy": 0.8390729090198874,
|
| 804 |
+
"epoch": 0.928,
|
| 805 |
+
"grad_norm": 19.25,
|
| 806 |
+
"learning_rate": 1.7933333333333333e-05,
|
| 807 |
+
"loss": 13.5587,
|
| 808 |
+
"mean_token_accuracy": 0.7872791964560747,
|
| 809 |
+
"num_tokens": 16499842.0,
|
| 810 |
+
"step": 580
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"entropy": 0.865405573323369,
|
| 814 |
+
"epoch": 0.944,
|
| 815 |
+
"grad_norm": 31.375,
|
| 816 |
+
"learning_rate": 1.785925925925926e-05,
|
| 817 |
+
"loss": 13.9929,
|
| 818 |
+
"mean_token_accuracy": 0.7827581968158484,
|
| 819 |
+
"num_tokens": 16781882.0,
|
| 820 |
+
"step": 590
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"entropy": 0.8798814173787832,
|
| 824 |
+
"epoch": 0.96,
|
| 825 |
+
"grad_norm": 14.6875,
|
| 826 |
+
"learning_rate": 1.7785185185185186e-05,
|
| 827 |
+
"loss": 14.0639,
|
| 828 |
+
"mean_token_accuracy": 0.7827699087560177,
|
| 829 |
+
"num_tokens": 17067407.0,
|
| 830 |
+
"step": 600
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.96,
|
| 834 |
+
"eval_biology_entropy": 3.278426109313965,
|
| 835 |
+
"eval_biology_loss": 3.676957607269287,
|
| 836 |
+
"eval_biology_mean_token_accuracy": 0.4364860031604767,
|
| 837 |
+
"eval_biology_num_tokens": 17067407.0,
|
| 838 |
+
"eval_biology_runtime": 49.9669,
|
| 839 |
+
"eval_biology_samples_per_second": 10.007,
|
| 840 |
+
"eval_biology_steps_per_second": 2.502,
|
| 841 |
+
"step": 600
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.96,
|
| 845 |
+
"eval_chemistry_entropy": 1.857783148765564,
|
| 846 |
+
"eval_chemistry_loss": 1.8980321884155273,
|
| 847 |
+
"eval_chemistry_mean_token_accuracy": 0.6259854953289032,
|
| 848 |
+
"eval_chemistry_num_tokens": 17067407.0,
|
| 849 |
+
"eval_chemistry_runtime": 62.1468,
|
| 850 |
+
"eval_chemistry_samples_per_second": 8.045,
|
| 851 |
+
"eval_chemistry_steps_per_second": 2.011,
|
| 852 |
+
"step": 600
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.96,
|
| 856 |
+
"eval_math_entropy": 0.8647160024642945,
|
| 857 |
+
"eval_math_loss": 0.868224024772644,
|
| 858 |
+
"eval_math_mean_token_accuracy": 0.7833898453712463,
|
| 859 |
+
"eval_math_num_tokens": 17067407.0,
|
| 860 |
+
"eval_math_runtime": 63.5747,
|
| 861 |
+
"eval_math_samples_per_second": 7.865,
|
| 862 |
+
"eval_math_steps_per_second": 1.966,
|
| 863 |
+
"step": 600
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.96,
|
| 867 |
+
"eval_physics_entropy": 1.4752862091064454,
|
| 868 |
+
"eval_physics_loss": 1.513607144355774,
|
| 869 |
+
"eval_physics_mean_token_accuracy": 0.6863754806518555,
|
| 870 |
+
"eval_physics_num_tokens": 17067407.0,
|
| 871 |
+
"eval_physics_runtime": 72.354,
|
| 872 |
+
"eval_physics_samples_per_second": 6.91,
|
| 873 |
+
"eval_physics_steps_per_second": 1.728,
|
| 874 |
+
"step": 600
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"entropy": 0.8497918715700508,
|
| 878 |
+
"epoch": 0.976,
|
| 879 |
+
"grad_norm": 18.625,
|
| 880 |
+
"learning_rate": 1.7711111111111113e-05,
|
| 881 |
+
"loss": 13.6664,
|
| 882 |
+
"mean_token_accuracy": 0.7879329767078161,
|
| 883 |
+
"num_tokens": 17350994.0,
|
| 884 |
+
"step": 610
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"entropy": 0.86589151956141,
|
| 888 |
+
"epoch": 0.992,
|
| 889 |
+
"grad_norm": 16.25,
|
| 890 |
+
"learning_rate": 1.763703703703704e-05,
|
| 891 |
+
"loss": 13.9208,
|
| 892 |
+
"mean_token_accuracy": 0.7827646978199482,
|
| 893 |
+
"num_tokens": 17637514.0,
|
| 894 |
+
"step": 620
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"entropy": 0.8501669876277447,
|
| 898 |
+
"epoch": 1.008,
|
| 899 |
+
"grad_norm": 15.375,
|
| 900 |
+
"learning_rate": 1.7562962962962962e-05,
|
| 901 |
+
"loss": 13.5677,
|
| 902 |
+
"mean_token_accuracy": 0.7862402476370335,
|
| 903 |
+
"num_tokens": 17926570.0,
|
| 904 |
+
"step": 630
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"entropy": 0.8345263335853815,
|
| 908 |
+
"epoch": 1.024,
|
| 909 |
+
"grad_norm": 15.625,
|
| 910 |
+
"learning_rate": 1.7488888888888892e-05,
|
| 911 |
+
"loss": 13.3578,
|
| 912 |
+
"mean_token_accuracy": 0.7888704732060432,
|
| 913 |
+
"num_tokens": 18207652.0,
|
| 914 |
+
"step": 640
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"entropy": 0.8276458965614438,
|
| 918 |
+
"epoch": 1.04,
|
| 919 |
+
"grad_norm": 17.5,
|
| 920 |
+
"learning_rate": 1.7414814814814815e-05,
|
| 921 |
+
"loss": 13.3452,
|
| 922 |
+
"mean_token_accuracy": 0.7908481806516647,
|
| 923 |
+
"num_tokens": 18484931.0,
|
| 924 |
+
"step": 650
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"entropy": 0.8310236493125558,
|
| 928 |
+
"epoch": 1.056,
|
| 929 |
+
"grad_norm": 16.875,
|
| 930 |
+
"learning_rate": 1.7340740740740742e-05,
|
| 931 |
+
"loss": 13.5223,
|
| 932 |
+
"mean_token_accuracy": 0.7887609034776688,
|
| 933 |
+
"num_tokens": 18773457.0,
|
| 934 |
+
"step": 660
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"entropy": 0.840316267311573,
|
| 938 |
+
"epoch": 1.072,
|
| 939 |
+
"grad_norm": 16.625,
|
| 940 |
+
"learning_rate": 1.726666666666667e-05,
|
| 941 |
+
"loss": 13.4228,
|
| 942 |
+
"mean_token_accuracy": 0.7874431516975164,
|
| 943 |
+
"num_tokens": 19055365.0,
|
| 944 |
+
"step": 670
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"entropy": 0.8311731787398458,
|
| 948 |
+
"epoch": 1.088,
|
| 949 |
+
"grad_norm": 14.4375,
|
| 950 |
+
"learning_rate": 1.7192592592592595e-05,
|
| 951 |
+
"loss": 13.3971,
|
| 952 |
+
"mean_token_accuracy": 0.790545703843236,
|
| 953 |
+
"num_tokens": 19345730.0,
|
| 954 |
+
"step": 680
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"entropy": 0.8338565738871694,
|
| 958 |
+
"epoch": 1.104,
|
| 959 |
+
"grad_norm": 13.6875,
|
| 960 |
+
"learning_rate": 1.711851851851852e-05,
|
| 961 |
+
"loss": 13.39,
|
| 962 |
+
"mean_token_accuracy": 0.7901448987424373,
|
| 963 |
+
"num_tokens": 19637390.0,
|
| 964 |
+
"step": 690
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"entropy": 0.8102976515889168,
|
| 968 |
+
"epoch": 1.12,
|
| 969 |
+
"grad_norm": 17.5,
|
| 970 |
+
"learning_rate": 1.7044444444444445e-05,
|
| 971 |
+
"loss": 12.9856,
|
| 972 |
+
"mean_token_accuracy": 0.7944051265716553,
|
| 973 |
+
"num_tokens": 19923914.0,
|
| 974 |
+
"step": 700
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 1.12,
|
| 978 |
+
"eval_biology_entropy": 3.2230891466140745,
|
| 979 |
+
"eval_biology_loss": 3.734968423843384,
|
| 980 |
+
"eval_biology_mean_token_accuracy": 0.4355302617549896,
|
| 981 |
+
"eval_biology_num_tokens": 19923914.0,
|
| 982 |
+
"eval_biology_runtime": 49.8901,
|
| 983 |
+
"eval_biology_samples_per_second": 10.022,
|
| 984 |
+
"eval_biology_steps_per_second": 2.506,
|
| 985 |
+
"step": 700
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 1.12,
|
| 989 |
+
"eval_chemistry_entropy": 1.7908948783874512,
|
| 990 |
+
"eval_chemistry_loss": 1.90095853805542,
|
| 991 |
+
"eval_chemistry_mean_token_accuracy": 0.6279562847614288,
|
| 992 |
+
"eval_chemistry_num_tokens": 19923914.0,
|
| 993 |
+
"eval_chemistry_runtime": 61.9568,
|
| 994 |
+
"eval_chemistry_samples_per_second": 8.07,
|
| 995 |
+
"eval_chemistry_steps_per_second": 2.018,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.12,
|
| 1000 |
+
"eval_math_entropy": 0.8332763781547546,
|
| 1001 |
+
"eval_math_loss": 0.8580161333084106,
|
| 1002 |
+
"eval_math_mean_token_accuracy": 0.7852629780769348,
|
| 1003 |
+
"eval_math_num_tokens": 19923914.0,
|
| 1004 |
+
"eval_math_runtime": 63.5133,
|
| 1005 |
+
"eval_math_samples_per_second": 7.872,
|
| 1006 |
+
"eval_math_steps_per_second": 1.968,
|
| 1007 |
+
"step": 700
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 1.12,
|
| 1011 |
+
"eval_physics_entropy": 1.421106719970703,
|
| 1012 |
+
"eval_physics_loss": 1.504669189453125,
|
| 1013 |
+
"eval_physics_mean_token_accuracy": 0.687330924987793,
|
| 1014 |
+
"eval_physics_num_tokens": 19923914.0,
|
| 1015 |
+
"eval_physics_runtime": 72.2525,
|
| 1016 |
+
"eval_physics_samples_per_second": 6.92,
|
| 1017 |
+
"eval_physics_steps_per_second": 1.73,
|
| 1018 |
+
"step": 700
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"entropy": 0.8049196273088455,
|
| 1022 |
+
"epoch": 1.1360000000000001,
|
| 1023 |
+
"grad_norm": 16.125,
|
| 1024 |
+
"learning_rate": 1.697037037037037e-05,
|
| 1025 |
+
"loss": 13.0378,
|
| 1026 |
+
"mean_token_accuracy": 0.7929212123155593,
|
| 1027 |
+
"num_tokens": 20201892.0,
|
| 1028 |
+
"step": 710
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"entropy": 0.8303030036389828,
|
| 1032 |
+
"epoch": 1.152,
|
| 1033 |
+
"grad_norm": 17.75,
|
| 1034 |
+
"learning_rate": 1.6896296296296298e-05,
|
| 1035 |
+
"loss": 13.304,
|
| 1036 |
+
"mean_token_accuracy": 0.7919872298836708,
|
| 1037 |
+
"num_tokens": 20490282.0,
|
| 1038 |
+
"step": 720
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"entropy": 0.8234290273860096,
|
| 1042 |
+
"epoch": 1.168,
|
| 1043 |
+
"grad_norm": 14.6875,
|
| 1044 |
+
"learning_rate": 1.6822222222222224e-05,
|
| 1045 |
+
"loss": 13.2574,
|
| 1046 |
+
"mean_token_accuracy": 0.791878254711628,
|
| 1047 |
+
"num_tokens": 20785786.0,
|
| 1048 |
+
"step": 730
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"entropy": 0.8201817765831947,
|
| 1052 |
+
"epoch": 1.184,
|
| 1053 |
+
"grad_norm": 18.25,
|
| 1054 |
+
"learning_rate": 1.6748148148148147e-05,
|
| 1055 |
+
"loss": 13.1797,
|
| 1056 |
+
"mean_token_accuracy": 0.7926335010677577,
|
| 1057 |
+
"num_tokens": 21074205.0,
|
| 1058 |
+
"step": 740
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"entropy": 0.8103797230869532,
|
| 1062 |
+
"epoch": 1.2,
|
| 1063 |
+
"grad_norm": 17.75,
|
| 1064 |
+
"learning_rate": 1.6674074074074077e-05,
|
| 1065 |
+
"loss": 13.0375,
|
| 1066 |
+
"mean_token_accuracy": 0.7937572699040174,
|
| 1067 |
+
"num_tokens": 21369159.0,
|
| 1068 |
+
"step": 750
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"entropy": 0.8331582929939032,
|
| 1072 |
+
"epoch": 1.216,
|
| 1073 |
+
"grad_norm": 21.5,
|
| 1074 |
+
"learning_rate": 1.66e-05,
|
| 1075 |
+
"loss": 13.3733,
|
| 1076 |
+
"mean_token_accuracy": 0.7907441522926092,
|
| 1077 |
+
"num_tokens": 21649178.0,
|
| 1078 |
+
"step": 760
|
| 1079 |
+
},
|
| 1080 |
+
{
|
| 1081 |
+
"entropy": 0.8037656577304005,
|
| 1082 |
+
"epoch": 1.232,
|
| 1083 |
+
"grad_norm": 17.125,
|
| 1084 |
+
"learning_rate": 1.6525925925925927e-05,
|
| 1085 |
+
"loss": 12.8855,
|
| 1086 |
+
"mean_token_accuracy": 0.7958086896687746,
|
| 1087 |
+
"num_tokens": 21930239.0,
|
| 1088 |
+
"step": 770
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"entropy": 0.7982386674731969,
|
| 1092 |
+
"epoch": 1.248,
|
| 1093 |
+
"grad_norm": 15.625,
|
| 1094 |
+
"learning_rate": 1.6451851851851853e-05,
|
| 1095 |
+
"loss": 12.8508,
|
| 1096 |
+
"mean_token_accuracy": 0.7953327149152756,
|
| 1097 |
+
"num_tokens": 22216387.0,
|
| 1098 |
+
"step": 780
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"entropy": 0.8371293131262064,
|
| 1102 |
+
"epoch": 1.264,
|
| 1103 |
+
"grad_norm": 18.625,
|
| 1104 |
+
"learning_rate": 1.637777777777778e-05,
|
| 1105 |
+
"loss": 13.5088,
|
| 1106 |
+
"mean_token_accuracy": 0.7873477075248957,
|
| 1107 |
+
"num_tokens": 22501002.0,
|
| 1108 |
+
"step": 790
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"entropy": 0.8327024061232805,
|
| 1112 |
+
"epoch": 1.28,
|
| 1113 |
+
"grad_norm": 20.625,
|
| 1114 |
+
"learning_rate": 1.6303703703703706e-05,
|
| 1115 |
+
"loss": 13.3064,
|
| 1116 |
+
"mean_token_accuracy": 0.791947865113616,
|
| 1117 |
+
"num_tokens": 22779682.0,
|
| 1118 |
+
"step": 800
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.28,
|
| 1122 |
+
"eval_biology_entropy": 3.294544553756714,
|
| 1123 |
+
"eval_biology_loss": 3.6706202030181885,
|
| 1124 |
+
"eval_biology_mean_token_accuracy": 0.43737580394744874,
|
| 1125 |
+
"eval_biology_num_tokens": 22779682.0,
|
| 1126 |
+
"eval_biology_runtime": 49.9678,
|
| 1127 |
+
"eval_biology_samples_per_second": 10.006,
|
| 1128 |
+
"eval_biology_steps_per_second": 2.502,
|
| 1129 |
+
"step": 800
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 1.28,
|
| 1133 |
+
"eval_chemistry_entropy": 1.8024000110626222,
|
| 1134 |
+
"eval_chemistry_loss": 1.8754929304122925,
|
| 1135 |
+
"eval_chemistry_mean_token_accuracy": 0.6302983772754669,
|
| 1136 |
+
"eval_chemistry_num_tokens": 22779682.0,
|
| 1137 |
+
"eval_chemistry_runtime": 62.133,
|
| 1138 |
+
"eval_chemistry_samples_per_second": 8.047,
|
| 1139 |
+
"eval_chemistry_steps_per_second": 2.012,
|
| 1140 |
+
"step": 800
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 1.28,
|
| 1144 |
+
"eval_math_entropy": 0.8196169562339782,
|
| 1145 |
+
"eval_math_loss": 0.8479236364364624,
|
| 1146 |
+
"eval_math_mean_token_accuracy": 0.7868858284950256,
|
| 1147 |
+
"eval_math_num_tokens": 22779682.0,
|
| 1148 |
+
"eval_math_runtime": 64.1319,
|
| 1149 |
+
"eval_math_samples_per_second": 7.796,
|
| 1150 |
+
"eval_math_steps_per_second": 1.949,
|
| 1151 |
+
"step": 800
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.28,
|
| 1155 |
+
"eval_physics_entropy": 1.4233626689910888,
|
| 1156 |
+
"eval_physics_loss": 1.4852280616760254,
|
| 1157 |
+
"eval_physics_mean_token_accuracy": 0.6896569218635559,
|
| 1158 |
+
"eval_physics_num_tokens": 22779682.0,
|
| 1159 |
+
"eval_physics_runtime": 72.356,
|
| 1160 |
+
"eval_physics_samples_per_second": 6.91,
|
| 1161 |
+
"eval_physics_steps_per_second": 1.728,
|
| 1162 |
+
"step": 800
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"entropy": 0.8020031837746501,
|
| 1166 |
+
"epoch": 1.296,
|
| 1167 |
+
"grad_norm": 16.25,
|
| 1168 |
+
"learning_rate": 1.622962962962963e-05,
|
| 1169 |
+
"loss": 12.998,
|
| 1170 |
+
"mean_token_accuracy": 0.7940779849886894,
|
| 1171 |
+
"num_tokens": 23057744.0,
|
| 1172 |
+
"step": 810
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"entropy": 0.812518722563982,
|
| 1176 |
+
"epoch": 1.312,
|
| 1177 |
+
"grad_norm": 19.125,
|
| 1178 |
+
"learning_rate": 1.6155555555555556e-05,
|
| 1179 |
+
"loss": 13.2282,
|
| 1180 |
+
"mean_token_accuracy": 0.7920650199055672,
|
| 1181 |
+
"num_tokens": 23344644.0,
|
| 1182 |
+
"step": 820
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"entropy": 0.8391162436455488,
|
| 1186 |
+
"epoch": 1.328,
|
| 1187 |
+
"grad_norm": 17.875,
|
| 1188 |
+
"learning_rate": 1.6081481481481482e-05,
|
| 1189 |
+
"loss": 13.3581,
|
| 1190 |
+
"mean_token_accuracy": 0.7894898921251297,
|
| 1191 |
+
"num_tokens": 23622405.0,
|
| 1192 |
+
"step": 830
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"entropy": 0.8036098016425968,
|
| 1196 |
+
"epoch": 1.3439999999999999,
|
| 1197 |
+
"grad_norm": 15.6875,
|
| 1198 |
+
"learning_rate": 1.600740740740741e-05,
|
| 1199 |
+
"loss": 12.9155,
|
| 1200 |
+
"mean_token_accuracy": 0.7945138659328222,
|
| 1201 |
+
"num_tokens": 23899771.0,
|
| 1202 |
+
"step": 840
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"entropy": 0.802362441085279,
|
| 1206 |
+
"epoch": 1.3599999999999999,
|
| 1207 |
+
"grad_norm": 14.5,
|
| 1208 |
+
"learning_rate": 1.5933333333333336e-05,
|
| 1209 |
+
"loss": 12.9723,
|
| 1210 |
+
"mean_token_accuracy": 0.7945049315690994,
|
| 1211 |
+
"num_tokens": 24187023.0,
|
| 1212 |
+
"step": 850
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"entropy": 0.8688682168722153,
|
| 1216 |
+
"epoch": 1.376,
|
| 1217 |
+
"grad_norm": 19.75,
|
| 1218 |
+
"learning_rate": 1.5859259259259262e-05,
|
| 1219 |
+
"loss": 13.9796,
|
| 1220 |
+
"mean_token_accuracy": 0.7812775958329439,
|
| 1221 |
+
"num_tokens": 24466132.0,
|
| 1222 |
+
"step": 860
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"entropy": 0.8100431380793452,
|
| 1226 |
+
"epoch": 1.392,
|
| 1227 |
+
"grad_norm": 18.75,
|
| 1228 |
+
"learning_rate": 1.5785185185185185e-05,
|
| 1229 |
+
"loss": 13.0219,
|
| 1230 |
+
"mean_token_accuracy": 0.7929740186780692,
|
| 1231 |
+
"num_tokens": 24748043.0,
|
| 1232 |
+
"step": 870
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"entropy": 0.7973114985972642,
|
| 1236 |
+
"epoch": 1.408,
|
| 1237 |
+
"grad_norm": 18.5,
|
| 1238 |
+
"learning_rate": 1.571111111111111e-05,
|
| 1239 |
+
"loss": 12.8964,
|
| 1240 |
+
"mean_token_accuracy": 0.7961425919085741,
|
| 1241 |
+
"num_tokens": 25036674.0,
|
| 1242 |
+
"step": 880
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"entropy": 0.7954017139971257,
|
| 1246 |
+
"epoch": 1.424,
|
| 1247 |
+
"grad_norm": 13.6875,
|
| 1248 |
+
"learning_rate": 1.5637037037037038e-05,
|
| 1249 |
+
"loss": 12.8143,
|
| 1250 |
+
"mean_token_accuracy": 0.7969338528811931,
|
| 1251 |
+
"num_tokens": 25324579.0,
|
| 1252 |
+
"step": 890
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"entropy": 0.8209576582536101,
|
| 1256 |
+
"epoch": 1.44,
|
| 1257 |
+
"grad_norm": 18.0,
|
| 1258 |
+
"learning_rate": 1.5562962962962965e-05,
|
| 1259 |
+
"loss": 13.2807,
|
| 1260 |
+
"mean_token_accuracy": 0.7910069234669208,
|
| 1261 |
+
"num_tokens": 25606824.0,
|
| 1262 |
+
"step": 900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 1.44,
|
| 1266 |
+
"eval_biology_entropy": 3.271074831008911,
|
| 1267 |
+
"eval_biology_loss": 3.627021074295044,
|
| 1268 |
+
"eval_biology_mean_token_accuracy": 0.43953498554229736,
|
| 1269 |
+
"eval_biology_num_tokens": 25606824.0,
|
| 1270 |
+
"eval_biology_runtime": 49.8927,
|
| 1271 |
+
"eval_biology_samples_per_second": 10.021,
|
| 1272 |
+
"eval_biology_steps_per_second": 2.505,
|
| 1273 |
+
"step": 900
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 1.44,
|
| 1277 |
+
"eval_chemistry_entropy": 1.8041245365142822,
|
| 1278 |
+
"eval_chemistry_loss": 1.8709218502044678,
|
| 1279 |
+
"eval_chemistry_mean_token_accuracy": 0.6310126931667328,
|
| 1280 |
+
"eval_chemistry_num_tokens": 25606824.0,
|
| 1281 |
+
"eval_chemistry_runtime": 62.0005,
|
| 1282 |
+
"eval_chemistry_samples_per_second": 8.064,
|
| 1283 |
+
"eval_chemistry_steps_per_second": 2.016,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.44,
|
| 1288 |
+
"eval_math_entropy": 0.8287842729091645,
|
| 1289 |
+
"eval_math_loss": 0.8401960134506226,
|
| 1290 |
+
"eval_math_mean_token_accuracy": 0.7886354041099548,
|
| 1291 |
+
"eval_math_num_tokens": 25606824.0,
|
| 1292 |
+
"eval_math_runtime": 63.5552,
|
| 1293 |
+
"eval_math_samples_per_second": 7.867,
|
| 1294 |
+
"eval_math_steps_per_second": 1.967,
|
| 1295 |
+
"step": 900
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 1.44,
|
| 1299 |
+
"eval_physics_entropy": 1.4311670818328857,
|
| 1300 |
+
"eval_physics_loss": 1.4744281768798828,
|
| 1301 |
+
"eval_physics_mean_token_accuracy": 0.6918323221206665,
|
| 1302 |
+
"eval_physics_num_tokens": 25606824.0,
|
| 1303 |
+
"eval_physics_runtime": 72.5619,
|
| 1304 |
+
"eval_physics_samples_per_second": 6.891,
|
| 1305 |
+
"eval_physics_steps_per_second": 1.723,
|
| 1306 |
+
"step": 900
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"entropy": 0.8060221856459975,
|
| 1310 |
+
"epoch": 1.456,
|
| 1311 |
+
"grad_norm": 14.8125,
|
| 1312 |
+
"learning_rate": 1.548888888888889e-05,
|
| 1313 |
+
"loss": 12.8502,
|
| 1314 |
+
"mean_token_accuracy": 0.7942074920982123,
|
| 1315 |
+
"num_tokens": 25886396.0,
|
| 1316 |
+
"step": 910
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"entropy": 0.804918241687119,
|
| 1320 |
+
"epoch": 1.472,
|
| 1321 |
+
"grad_norm": 16.5,
|
| 1322 |
+
"learning_rate": 1.5414814814814814e-05,
|
| 1323 |
+
"loss": 12.9431,
|
| 1324 |
+
"mean_token_accuracy": 0.794098413363099,
|
| 1325 |
+
"num_tokens": 26163618.0,
|
| 1326 |
+
"step": 920
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"entropy": 0.8308094689622522,
|
| 1330 |
+
"epoch": 1.488,
|
| 1331 |
+
"grad_norm": 17.375,
|
| 1332 |
+
"learning_rate": 1.5340740740740744e-05,
|
| 1333 |
+
"loss": 13.5052,
|
| 1334 |
+
"mean_token_accuracy": 0.7875379234552383,
|
| 1335 |
+
"num_tokens": 26438338.0,
|
| 1336 |
+
"step": 930
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"entropy": 0.7895203305408358,
|
| 1340 |
+
"epoch": 1.504,
|
| 1341 |
+
"grad_norm": 20.25,
|
| 1342 |
+
"learning_rate": 1.5266666666666667e-05,
|
| 1343 |
+
"loss": 12.7945,
|
| 1344 |
+
"mean_token_accuracy": 0.796988645941019,
|
| 1345 |
+
"num_tokens": 26729255.0,
|
| 1346 |
+
"step": 940
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"entropy": 0.8357090651988983,
|
| 1350 |
+
"epoch": 1.52,
|
| 1351 |
+
"grad_norm": 15.5,
|
| 1352 |
+
"learning_rate": 1.5192592592592594e-05,
|
| 1353 |
+
"loss": 13.3142,
|
| 1354 |
+
"mean_token_accuracy": 0.790850094333291,
|
| 1355 |
+
"num_tokens": 27017935.0,
|
| 1356 |
+
"step": 950
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"entropy": 0.8033582132309676,
|
| 1360 |
+
"epoch": 1.536,
|
| 1361 |
+
"grad_norm": 14.8125,
|
| 1362 |
+
"learning_rate": 1.5118518518518519e-05,
|
| 1363 |
+
"loss": 12.9716,
|
| 1364 |
+
"mean_token_accuracy": 0.7936928037554025,
|
| 1365 |
+
"num_tokens": 27306339.0,
|
| 1366 |
+
"step": 960
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"entropy": 0.8104204999282956,
|
| 1370 |
+
"epoch": 1.552,
|
| 1371 |
+
"grad_norm": 16.625,
|
| 1372 |
+
"learning_rate": 1.5044444444444445e-05,
|
| 1373 |
+
"loss": 12.981,
|
| 1374 |
+
"mean_token_accuracy": 0.7937849141657353,
|
| 1375 |
+
"num_tokens": 27591959.0,
|
| 1376 |
+
"step": 970
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"entropy": 0.7846602015197277,
|
| 1380 |
+
"epoch": 1.568,
|
| 1381 |
+
"grad_norm": 14.0625,
|
| 1382 |
+
"learning_rate": 1.497037037037037e-05,
|
| 1383 |
+
"loss": 12.677,
|
| 1384 |
+
"mean_token_accuracy": 0.7991001665592193,
|
| 1385 |
+
"num_tokens": 27884398.0,
|
| 1386 |
+
"step": 980
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"entropy": 0.7992334404960275,
|
| 1390 |
+
"epoch": 1.584,
|
| 1391 |
+
"grad_norm": 18.875,
|
| 1392 |
+
"learning_rate": 1.4896296296296298e-05,
|
| 1393 |
+
"loss": 12.935,
|
| 1394 |
+
"mean_token_accuracy": 0.794677285850048,
|
| 1395 |
+
"num_tokens": 28171274.0,
|
| 1396 |
+
"step": 990
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"entropy": 0.8037812670692801,
|
| 1400 |
+
"epoch": 1.6,
|
| 1401 |
+
"grad_norm": 19.5,
|
| 1402 |
+
"learning_rate": 1.4822222222222225e-05,
|
| 1403 |
+
"loss": 12.7702,
|
| 1404 |
+
"mean_token_accuracy": 0.7950391560792923,
|
| 1405 |
+
"num_tokens": 28457624.0,
|
| 1406 |
+
"step": 1000
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 1.6,
|
| 1410 |
+
"eval_biology_entropy": 3.2400936613082885,
|
| 1411 |
+
"eval_biology_loss": 3.704118251800537,
|
| 1412 |
+
"eval_biology_mean_token_accuracy": 0.43956737661361694,
|
| 1413 |
+
"eval_biology_num_tokens": 28457624.0,
|
| 1414 |
+
"eval_biology_runtime": 49.8865,
|
| 1415 |
+
"eval_biology_samples_per_second": 10.023,
|
| 1416 |
+
"eval_biology_steps_per_second": 2.506,
|
| 1417 |
+
"step": 1000
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.6,
|
| 1421 |
+
"eval_chemistry_entropy": 1.776292507648468,
|
| 1422 |
+
"eval_chemistry_loss": 1.871563196182251,
|
| 1423 |
+
"eval_chemistry_mean_token_accuracy": 0.632093154668808,
|
| 1424 |
+
"eval_chemistry_num_tokens": 28457624.0,
|
| 1425 |
+
"eval_chemistry_runtime": 61.9067,
|
| 1426 |
+
"eval_chemistry_samples_per_second": 8.077,
|
| 1427 |
+
"eval_chemistry_steps_per_second": 2.019,
|
| 1428 |
+
"step": 1000
|
| 1429 |
+
},
|
| 1430 |
+
{
|
| 1431 |
+
"epoch": 1.6,
|
| 1432 |
+
"eval_math_entropy": 0.8142018506526947,
|
| 1433 |
+
"eval_math_loss": 0.834656834602356,
|
| 1434 |
+
"eval_math_mean_token_accuracy": 0.7897459998130798,
|
| 1435 |
+
"eval_math_num_tokens": 28457624.0,
|
| 1436 |
+
"eval_math_runtime": 64.1472,
|
| 1437 |
+
"eval_math_samples_per_second": 7.795,
|
| 1438 |
+
"eval_math_steps_per_second": 1.949,
|
| 1439 |
+
"step": 1000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 1.6,
|
| 1443 |
+
"eval_physics_entropy": 1.403171688079834,
|
| 1444 |
+
"eval_physics_loss": 1.471365213394165,
|
| 1445 |
+
"eval_physics_mean_token_accuracy": 0.692293493270874,
|
| 1446 |
+
"eval_physics_num_tokens": 28457624.0,
|
| 1447 |
+
"eval_physics_runtime": 72.7239,
|
| 1448 |
+
"eval_physics_samples_per_second": 6.875,
|
| 1449 |
+
"eval_physics_steps_per_second": 1.719,
|
| 1450 |
+
"step": 1000
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"entropy": 0.798175222799182,
|
| 1454 |
+
"epoch": 1.616,
|
| 1455 |
+
"grad_norm": 15.0,
|
| 1456 |
+
"learning_rate": 1.474814814814815e-05,
|
| 1457 |
+
"loss": 12.9819,
|
| 1458 |
+
"mean_token_accuracy": 0.7934872336685658,
|
| 1459 |
+
"num_tokens": 28743099.0,
|
| 1460 |
+
"step": 1010
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"entropy": 0.8180070934817195,
|
| 1464 |
+
"epoch": 1.6320000000000001,
|
| 1465 |
+
"grad_norm": 16.5,
|
| 1466 |
+
"learning_rate": 1.4674074074074076e-05,
|
| 1467 |
+
"loss": 13.0308,
|
| 1468 |
+
"mean_token_accuracy": 0.7923920370638371,
|
| 1469 |
+
"num_tokens": 29017297.0,
|
| 1470 |
+
"step": 1020
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"entropy": 0.7903204390779137,
|
| 1474 |
+
"epoch": 1.6480000000000001,
|
| 1475 |
+
"grad_norm": 17.5,
|
| 1476 |
+
"learning_rate": 1.46e-05,
|
| 1477 |
+
"loss": 12.7676,
|
| 1478 |
+
"mean_token_accuracy": 0.7979931388050318,
|
| 1479 |
+
"num_tokens": 29303707.0,
|
| 1480 |
+
"step": 1030
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"entropy": 0.7956463107839227,
|
| 1484 |
+
"epoch": 1.6640000000000001,
|
| 1485 |
+
"grad_norm": 16.5,
|
| 1486 |
+
"learning_rate": 1.4525925925925927e-05,
|
| 1487 |
+
"loss": 12.793,
|
| 1488 |
+
"mean_token_accuracy": 0.7947848867624998,
|
| 1489 |
+
"num_tokens": 29597156.0,
|
| 1490 |
+
"step": 1040
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"entropy": 0.8131252270191908,
|
| 1494 |
+
"epoch": 1.6800000000000002,
|
| 1495 |
+
"grad_norm": 15.3125,
|
| 1496 |
+
"learning_rate": 1.4451851851851852e-05,
|
| 1497 |
+
"loss": 13.0555,
|
| 1498 |
+
"mean_token_accuracy": 0.7951273001730442,
|
| 1499 |
+
"num_tokens": 29883879.0,
|
| 1500 |
+
"step": 1050
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"entropy": 0.8136409839615226,
|
| 1504 |
+
"epoch": 1.696,
|
| 1505 |
+
"grad_norm": 17.625,
|
| 1506 |
+
"learning_rate": 1.4377777777777779e-05,
|
| 1507 |
+
"loss": 13.173,
|
| 1508 |
+
"mean_token_accuracy": 0.7925207346677781,
|
| 1509 |
+
"num_tokens": 30165760.0,
|
| 1510 |
+
"step": 1060
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"entropy": 0.7853763310238719,
|
| 1514 |
+
"epoch": 1.712,
|
| 1515 |
+
"grad_norm": 16.125,
|
| 1516 |
+
"learning_rate": 1.4303703703703703e-05,
|
| 1517 |
+
"loss": 12.7266,
|
| 1518 |
+
"mean_token_accuracy": 0.7975193675607443,
|
| 1519 |
+
"num_tokens": 30460367.0,
|
| 1520 |
+
"step": 1070
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"entropy": 0.7903000565245748,
|
| 1524 |
+
"epoch": 1.728,
|
| 1525 |
+
"grad_norm": 21.625,
|
| 1526 |
+
"learning_rate": 1.4229629629629632e-05,
|
| 1527 |
+
"loss": 12.6329,
|
| 1528 |
+
"mean_token_accuracy": 0.7987389255315065,
|
| 1529 |
+
"num_tokens": 30739137.0,
|
| 1530 |
+
"step": 1080
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"entropy": 0.8021053321659565,
|
| 1534 |
+
"epoch": 1.744,
|
| 1535 |
+
"grad_norm": 17.625,
|
| 1536 |
+
"learning_rate": 1.4155555555555556e-05,
|
| 1537 |
+
"loss": 12.834,
|
| 1538 |
+
"mean_token_accuracy": 0.7961107190698385,
|
| 1539 |
+
"num_tokens": 31022663.0,
|
| 1540 |
+
"step": 1090
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"entropy": 0.7864446669816971,
|
| 1544 |
+
"epoch": 1.76,
|
| 1545 |
+
"grad_norm": 16.625,
|
| 1546 |
+
"learning_rate": 1.4081481481481483e-05,
|
| 1547 |
+
"loss": 12.6971,
|
| 1548 |
+
"mean_token_accuracy": 0.7962396703660488,
|
| 1549 |
+
"num_tokens": 31306494.0,
|
| 1550 |
+
"step": 1100
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.76,
|
| 1554 |
+
"eval_biology_entropy": 3.232375309944153,
|
| 1555 |
+
"eval_biology_loss": 3.648139238357544,
|
| 1556 |
+
"eval_biology_mean_token_accuracy": 0.4412148265838623,
|
| 1557 |
+
"eval_biology_num_tokens": 31306494.0,
|
| 1558 |
+
"eval_biology_runtime": 49.9213,
|
| 1559 |
+
"eval_biology_samples_per_second": 10.016,
|
| 1560 |
+
"eval_biology_steps_per_second": 2.504,
|
| 1561 |
+
"step": 1100
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.76,
|
| 1565 |
+
"eval_chemistry_entropy": 1.762632459640503,
|
| 1566 |
+
"eval_chemistry_loss": 1.86100435256958,
|
| 1567 |
+
"eval_chemistry_mean_token_accuracy": 0.6327439775466919,
|
| 1568 |
+
"eval_chemistry_num_tokens": 31306494.0,
|
| 1569 |
+
"eval_chemistry_runtime": 61.921,
|
| 1570 |
+
"eval_chemistry_samples_per_second": 8.075,
|
| 1571 |
+
"eval_chemistry_steps_per_second": 2.019,
|
| 1572 |
+
"step": 1100
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 1.76,
|
| 1576 |
+
"eval_math_entropy": 0.8122558786869049,
|
| 1577 |
+
"eval_math_loss": 0.8297922015190125,
|
| 1578 |
+
"eval_math_mean_token_accuracy": 0.7903609256744385,
|
| 1579 |
+
"eval_math_num_tokens": 31306494.0,
|
| 1580 |
+
"eval_math_runtime": 63.4704,
|
| 1581 |
+
"eval_math_samples_per_second": 7.878,
|
| 1582 |
+
"eval_math_steps_per_second": 1.969,
|
| 1583 |
+
"step": 1100
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 1.76,
|
| 1587 |
+
"eval_physics_entropy": 1.4011106457710265,
|
| 1588 |
+
"eval_physics_loss": 1.4673256874084473,
|
| 1589 |
+
"eval_physics_mean_token_accuracy": 0.692846088886261,
|
| 1590 |
+
"eval_physics_num_tokens": 31306494.0,
|
| 1591 |
+
"eval_physics_runtime": 72.2185,
|
| 1592 |
+
"eval_physics_samples_per_second": 6.923,
|
| 1593 |
+
"eval_physics_steps_per_second": 1.731,
|
| 1594 |
+
"step": 1100
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"entropy": 0.8044054295867682,
|
| 1598 |
+
"epoch": 1.776,
|
| 1599 |
+
"grad_norm": 17.125,
|
| 1600 |
+
"learning_rate": 1.400740740740741e-05,
|
| 1601 |
+
"loss": 12.9493,
|
| 1602 |
+
"mean_token_accuracy": 0.7960755385458469,
|
| 1603 |
+
"num_tokens": 31595542.0,
|
| 1604 |
+
"step": 1110
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"entropy": 0.8096846660599113,
|
| 1608 |
+
"epoch": 1.792,
|
| 1609 |
+
"grad_norm": 16.125,
|
| 1610 |
+
"learning_rate": 1.3933333333333334e-05,
|
| 1611 |
+
"loss": 13.0798,
|
| 1612 |
+
"mean_token_accuracy": 0.7921652976423502,
|
| 1613 |
+
"num_tokens": 31881189.0,
|
| 1614 |
+
"step": 1120
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"entropy": 0.8158628990873694,
|
| 1618 |
+
"epoch": 1.808,
|
| 1619 |
+
"grad_norm": 15.1875,
|
| 1620 |
+
"learning_rate": 1.385925925925926e-05,
|
| 1621 |
+
"loss": 12.9956,
|
| 1622 |
+
"mean_token_accuracy": 0.7941187128424645,
|
| 1623 |
+
"num_tokens": 32164196.0,
|
| 1624 |
+
"step": 1130
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"entropy": 0.7536683512851596,
|
| 1628 |
+
"epoch": 1.8239999999999998,
|
| 1629 |
+
"grad_norm": 15.75,
|
| 1630 |
+
"learning_rate": 1.3785185185185186e-05,
|
| 1631 |
+
"loss": 12.2054,
|
| 1632 |
+
"mean_token_accuracy": 0.8031131632626056,
|
| 1633 |
+
"num_tokens": 32441530.0,
|
| 1634 |
+
"step": 1140
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"entropy": 0.7897313551977276,
|
| 1638 |
+
"epoch": 1.8399999999999999,
|
| 1639 |
+
"grad_norm": 18.625,
|
| 1640 |
+
"learning_rate": 1.3711111111111112e-05,
|
| 1641 |
+
"loss": 12.7218,
|
| 1642 |
+
"mean_token_accuracy": 0.7971313249319791,
|
| 1643 |
+
"num_tokens": 32723145.0,
|
| 1644 |
+
"step": 1150
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"entropy": 0.8171826468780636,
|
| 1648 |
+
"epoch": 1.8559999999999999,
|
| 1649 |
+
"grad_norm": 15.125,
|
| 1650 |
+
"learning_rate": 1.3637037037037037e-05,
|
| 1651 |
+
"loss": 13.0642,
|
| 1652 |
+
"mean_token_accuracy": 0.7934288665652275,
|
| 1653 |
+
"num_tokens": 33011263.0,
|
| 1654 |
+
"step": 1160
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"entropy": 0.7859557889401912,
|
| 1658 |
+
"epoch": 1.8719999999999999,
|
| 1659 |
+
"grad_norm": 15.75,
|
| 1660 |
+
"learning_rate": 1.3562962962962965e-05,
|
| 1661 |
+
"loss": 12.6724,
|
| 1662 |
+
"mean_token_accuracy": 0.8001265831291675,
|
| 1663 |
+
"num_tokens": 33298921.0,
|
| 1664 |
+
"step": 1170
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"entropy": 0.8064542135223747,
|
| 1668 |
+
"epoch": 1.888,
|
| 1669 |
+
"grad_norm": 18.125,
|
| 1670 |
+
"learning_rate": 1.3488888888888888e-05,
|
| 1671 |
+
"loss": 12.9109,
|
| 1672 |
+
"mean_token_accuracy": 0.7947272609919309,
|
| 1673 |
+
"num_tokens": 33576243.0,
|
| 1674 |
+
"step": 1180
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"entropy": 0.7957150738686323,
|
| 1678 |
+
"epoch": 1.904,
|
| 1679 |
+
"grad_norm": 15.375,
|
| 1680 |
+
"learning_rate": 1.3414814814814817e-05,
|
| 1681 |
+
"loss": 12.8458,
|
| 1682 |
+
"mean_token_accuracy": 0.7957113239914179,
|
| 1683 |
+
"num_tokens": 33850968.0,
|
| 1684 |
+
"step": 1190
|
| 1685 |
+
},
|
| 1686 |
+
{
|
| 1687 |
+
"entropy": 0.7726223962381482,
|
| 1688 |
+
"epoch": 1.92,
|
| 1689 |
+
"grad_norm": 15.125,
|
| 1690 |
+
"learning_rate": 1.3340740740740741e-05,
|
| 1691 |
+
"loss": 12.4511,
|
| 1692 |
+
"mean_token_accuracy": 0.8002966489642859,
|
| 1693 |
+
"num_tokens": 34128558.0,
|
| 1694 |
+
"step": 1200
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 1.92,
|
| 1698 |
+
"eval_biology_entropy": 3.183071039199829,
|
| 1699 |
+
"eval_biology_loss": 3.584599494934082,
|
| 1700 |
+
"eval_biology_mean_token_accuracy": 0.44353048944473267,
|
| 1701 |
+
"eval_biology_num_tokens": 34128558.0,
|
| 1702 |
+
"eval_biology_runtime": 50.2602,
|
| 1703 |
+
"eval_biology_samples_per_second": 9.948,
|
| 1704 |
+
"eval_biology_steps_per_second": 2.487,
|
| 1705 |
+
"step": 1200
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 1.92,
|
| 1709 |
+
"eval_chemistry_entropy": 1.742585494041443,
|
| 1710 |
+
"eval_chemistry_loss": 1.850415825843811,
|
| 1711 |
+
"eval_chemistry_mean_token_accuracy": 0.6337505750656128,
|
| 1712 |
+
"eval_chemistry_num_tokens": 34128558.0,
|
| 1713 |
+
"eval_chemistry_runtime": 61.9178,
|
| 1714 |
+
"eval_chemistry_samples_per_second": 8.075,
|
| 1715 |
+
"eval_chemistry_steps_per_second": 2.019,
|
| 1716 |
+
"step": 1200
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 1.92,
|
| 1720 |
+
"eval_math_entropy": 0.812834040403366,
|
| 1721 |
+
"eval_math_loss": 0.8256640434265137,
|
| 1722 |
+
"eval_math_mean_token_accuracy": 0.7911388430595397,
|
| 1723 |
+
"eval_math_num_tokens": 34128558.0,
|
| 1724 |
+
"eval_math_runtime": 63.4908,
|
| 1725 |
+
"eval_math_samples_per_second": 7.875,
|
| 1726 |
+
"eval_math_steps_per_second": 1.969,
|
| 1727 |
+
"step": 1200
|
| 1728 |
+
},
|
| 1729 |
+
{
|
| 1730 |
+
"epoch": 1.92,
|
| 1731 |
+
"eval_physics_entropy": 1.396484270095825,
|
| 1732 |
+
"eval_physics_loss": 1.4663584232330322,
|
| 1733 |
+
"eval_physics_mean_token_accuracy": 0.6928882389068604,
|
| 1734 |
+
"eval_physics_num_tokens": 34128558.0,
|
| 1735 |
+
"eval_physics_runtime": 72.4413,
|
| 1736 |
+
"eval_physics_samples_per_second": 6.902,
|
| 1737 |
+
"eval_physics_steps_per_second": 1.726,
|
| 1738 |
+
"step": 1200
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"entropy": 0.8009239070117473,
|
| 1742 |
+
"epoch": 1.936,
|
| 1743 |
+
"grad_norm": 21.125,
|
| 1744 |
+
"learning_rate": 1.3266666666666668e-05,
|
| 1745 |
+
"loss": 12.8074,
|
| 1746 |
+
"mean_token_accuracy": 0.7954005278646946,
|
| 1747 |
+
"num_tokens": 34408056.0,
|
| 1748 |
+
"step": 1210
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"entropy": 0.7883490238338708,
|
| 1752 |
+
"epoch": 1.952,
|
| 1753 |
+
"grad_norm": 18.75,
|
| 1754 |
+
"learning_rate": 1.3192592592592594e-05,
|
| 1755 |
+
"loss": 12.7364,
|
| 1756 |
+
"mean_token_accuracy": 0.7968561142683029,
|
| 1757 |
+
"num_tokens": 34684679.0,
|
| 1758 |
+
"step": 1220
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"entropy": 0.8134514577686787,
|
| 1762 |
+
"epoch": 1.968,
|
| 1763 |
+
"grad_norm": 17.625,
|
| 1764 |
+
"learning_rate": 1.311851851851852e-05,
|
| 1765 |
+
"loss": 13.2068,
|
| 1766 |
+
"mean_token_accuracy": 0.7917527597397566,
|
| 1767 |
+
"num_tokens": 34971038.0,
|
| 1768 |
+
"step": 1230
|
| 1769 |
+
},
|
| 1770 |
+
{
|
| 1771 |
+
"entropy": 0.7896538088098168,
|
| 1772 |
+
"epoch": 1.984,
|
| 1773 |
+
"grad_norm": 16.0,
|
| 1774 |
+
"learning_rate": 1.3044444444444446e-05,
|
| 1775 |
+
"loss": 12.6304,
|
| 1776 |
+
"mean_token_accuracy": 0.7988350588828326,
|
| 1777 |
+
"num_tokens": 35262281.0,
|
| 1778 |
+
"step": 1240
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"entropy": 0.7750496102496982,
|
| 1782 |
+
"epoch": 2.0,
|
| 1783 |
+
"grad_norm": 19.25,
|
| 1784 |
+
"learning_rate": 1.297037037037037e-05,
|
| 1785 |
+
"loss": 12.492,
|
| 1786 |
+
"mean_token_accuracy": 0.799928730726242,
|
| 1787 |
+
"num_tokens": 35560864.0,
|
| 1788 |
+
"step": 1250
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"entropy": 0.8006911057978868,
|
| 1792 |
+
"epoch": 2.016,
|
| 1793 |
+
"grad_norm": 14.1875,
|
| 1794 |
+
"learning_rate": 1.2896296296296299e-05,
|
| 1795 |
+
"loss": 12.8078,
|
| 1796 |
+
"mean_token_accuracy": 0.7973089322447777,
|
| 1797 |
+
"num_tokens": 35846704.0,
|
| 1798 |
+
"step": 1260
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"entropy": 0.7901067767292261,
|
| 1802 |
+
"epoch": 2.032,
|
| 1803 |
+
"grad_norm": 17.625,
|
| 1804 |
+
"learning_rate": 1.2822222222222222e-05,
|
| 1805 |
+
"loss": 12.6518,
|
| 1806 |
+
"mean_token_accuracy": 0.7975228931754827,
|
| 1807 |
+
"num_tokens": 36128775.0,
|
| 1808 |
+
"step": 1270
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"entropy": 0.7684864716604352,
|
| 1812 |
+
"epoch": 2.048,
|
| 1813 |
+
"grad_norm": 16.875,
|
| 1814 |
+
"learning_rate": 1.274814814814815e-05,
|
| 1815 |
+
"loss": 12.2821,
|
| 1816 |
+
"mean_token_accuracy": 0.803328375890851,
|
| 1817 |
+
"num_tokens": 36419504.0,
|
| 1818 |
+
"step": 1280
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"entropy": 0.7612387906759978,
|
| 1822 |
+
"epoch": 2.064,
|
| 1823 |
+
"grad_norm": 16.25,
|
| 1824 |
+
"learning_rate": 1.2674074074074075e-05,
|
| 1825 |
+
"loss": 12.3434,
|
| 1826 |
+
"mean_token_accuracy": 0.8015977006405592,
|
| 1827 |
+
"num_tokens": 36706816.0,
|
| 1828 |
+
"step": 1290
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"entropy": 0.7767295489087701,
|
| 1832 |
+
"epoch": 2.08,
|
| 1833 |
+
"grad_norm": 16.375,
|
| 1834 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 1835 |
+
"loss": 12.345,
|
| 1836 |
+
"mean_token_accuracy": 0.8008607290685177,
|
| 1837 |
+
"num_tokens": 36988475.0,
|
| 1838 |
+
"step": 1300
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 2.08,
|
| 1842 |
+
"eval_biology_entropy": 3.112719190597534,
|
| 1843 |
+
"eval_biology_loss": 3.577775716781616,
|
| 1844 |
+
"eval_biology_mean_token_accuracy": 0.4433496446609497,
|
| 1845 |
+
"eval_biology_num_tokens": 36988475.0,
|
| 1846 |
+
"eval_biology_runtime": 50.4638,
|
| 1847 |
+
"eval_biology_samples_per_second": 9.908,
|
| 1848 |
+
"eval_biology_steps_per_second": 2.477,
|
| 1849 |
+
"step": 1300
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 2.08,
|
| 1853 |
+
"eval_chemistry_entropy": 1.6892164297103882,
|
| 1854 |
+
"eval_chemistry_loss": 1.852262258529663,
|
| 1855 |
+
"eval_chemistry_mean_token_accuracy": 0.6334386568069458,
|
| 1856 |
+
"eval_chemistry_num_tokens": 36988475.0,
|
| 1857 |
+
"eval_chemistry_runtime": 62.4497,
|
| 1858 |
+
"eval_chemistry_samples_per_second": 8.006,
|
| 1859 |
+
"eval_chemistry_steps_per_second": 2.002,
|
| 1860 |
+
"step": 1300
|
| 1861 |
+
},
|
| 1862 |
+
{
|
| 1863 |
+
"epoch": 2.08,
|
| 1864 |
+
"eval_math_entropy": 0.7883058032989502,
|
| 1865 |
+
"eval_math_loss": 0.8236408829689026,
|
| 1866 |
+
"eval_math_mean_token_accuracy": 0.7912847075462341,
|
| 1867 |
+
"eval_math_num_tokens": 36988475.0,
|
| 1868 |
+
"eval_math_runtime": 63.9109,
|
| 1869 |
+
"eval_math_samples_per_second": 7.823,
|
| 1870 |
+
"eval_math_steps_per_second": 1.956,
|
| 1871 |
+
"step": 1300
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 2.08,
|
| 1875 |
+
"eval_physics_entropy": 1.3466296257972716,
|
| 1876 |
+
"eval_physics_loss": 1.459655523300171,
|
| 1877 |
+
"eval_physics_mean_token_accuracy": 0.6937613749504089,
|
| 1878 |
+
"eval_physics_num_tokens": 36988475.0,
|
| 1879 |
+
"eval_physics_runtime": 72.3423,
|
| 1880 |
+
"eval_physics_samples_per_second": 6.912,
|
| 1881 |
+
"eval_physics_steps_per_second": 1.728,
|
| 1882 |
+
"step": 1300
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"entropy": 0.7833418918773531,
|
| 1886 |
+
"epoch": 2.096,
|
| 1887 |
+
"grad_norm": 15.9375,
|
| 1888 |
+
"learning_rate": 1.2525925925925928e-05,
|
| 1889 |
+
"loss": 12.6724,
|
| 1890 |
+
"mean_token_accuracy": 0.7996709801256656,
|
| 1891 |
+
"num_tokens": 37270131.0,
|
| 1892 |
+
"step": 1310
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"entropy": 0.7611293364316225,
|
| 1896 |
+
"epoch": 2.112,
|
| 1897 |
+
"grad_norm": 15.5,
|
| 1898 |
+
"learning_rate": 1.2451851851851853e-05,
|
| 1899 |
+
"loss": 12.2516,
|
| 1900 |
+
"mean_token_accuracy": 0.8012572944164276,
|
| 1901 |
+
"num_tokens": 37563537.0,
|
| 1902 |
+
"step": 1320
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"entropy": 0.7628815559670329,
|
| 1906 |
+
"epoch": 2.128,
|
| 1907 |
+
"grad_norm": 18.375,
|
| 1908 |
+
"learning_rate": 1.237777777777778e-05,
|
| 1909 |
+
"loss": 12.157,
|
| 1910 |
+
"mean_token_accuracy": 0.8029138047248126,
|
| 1911 |
+
"num_tokens": 37843959.0,
|
| 1912 |
+
"step": 1330
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"entropy": 0.7627110643312335,
|
| 1916 |
+
"epoch": 2.144,
|
| 1917 |
+
"grad_norm": 16.875,
|
| 1918 |
+
"learning_rate": 1.2303703703703704e-05,
|
| 1919 |
+
"loss": 12.3203,
|
| 1920 |
+
"mean_token_accuracy": 0.8017950035631657,
|
| 1921 |
+
"num_tokens": 38133092.0,
|
| 1922 |
+
"step": 1340
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"entropy": 0.7732076404616237,
|
| 1926 |
+
"epoch": 2.16,
|
| 1927 |
+
"grad_norm": 15.5625,
|
| 1928 |
+
"learning_rate": 1.222962962962963e-05,
|
| 1929 |
+
"loss": 12.3963,
|
| 1930 |
+
"mean_token_accuracy": 0.8020700648427009,
|
| 1931 |
+
"num_tokens": 38421229.0,
|
| 1932 |
+
"step": 1350
|
| 1933 |
+
},
|
| 1934 |
+
{
|
| 1935 |
+
"entropy": 0.7416583737358451,
|
| 1936 |
+
"epoch": 2.176,
|
| 1937 |
+
"grad_norm": 19.375,
|
| 1938 |
+
"learning_rate": 1.2155555555555555e-05,
|
| 1939 |
+
"loss": 11.9055,
|
| 1940 |
+
"mean_token_accuracy": 0.8070738401263953,
|
| 1941 |
+
"num_tokens": 38708043.0,
|
| 1942 |
+
"step": 1360
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"entropy": 0.7689750671386719,
|
| 1946 |
+
"epoch": 2.192,
|
| 1947 |
+
"grad_norm": 15.9375,
|
| 1948 |
+
"learning_rate": 1.2081481481481484e-05,
|
| 1949 |
+
"loss": 12.3704,
|
| 1950 |
+
"mean_token_accuracy": 0.800269627571106,
|
| 1951 |
+
"num_tokens": 38996930.0,
|
| 1952 |
+
"step": 1370
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"entropy": 0.7766441306099295,
|
| 1956 |
+
"epoch": 2.208,
|
| 1957 |
+
"grad_norm": 29.375,
|
| 1958 |
+
"learning_rate": 1.2007407407407408e-05,
|
| 1959 |
+
"loss": 12.5879,
|
| 1960 |
+
"mean_token_accuracy": 0.7995092123746872,
|
| 1961 |
+
"num_tokens": 39279481.0,
|
| 1962 |
+
"step": 1380
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"entropy": 0.7676854202523827,
|
| 1966 |
+
"epoch": 2.224,
|
| 1967 |
+
"grad_norm": 18.375,
|
| 1968 |
+
"learning_rate": 1.1933333333333335e-05,
|
| 1969 |
+
"loss": 12.3774,
|
| 1970 |
+
"mean_token_accuracy": 0.8020657405257225,
|
| 1971 |
+
"num_tokens": 39569030.0,
|
| 1972 |
+
"step": 1390
|
| 1973 |
+
},
|
| 1974 |
+
{
|
| 1975 |
+
"entropy": 0.7721736744046211,
|
| 1976 |
+
"epoch": 2.24,
|
| 1977 |
+
"grad_norm": 16.5,
|
| 1978 |
+
"learning_rate": 1.185925925925926e-05,
|
| 1979 |
+
"loss": 12.385,
|
| 1980 |
+
"mean_token_accuracy": 0.8002489194273948,
|
| 1981 |
+
"num_tokens": 39854873.0,
|
| 1982 |
+
"step": 1400
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 2.24,
|
| 1986 |
+
"eval_biology_entropy": 3.1739475746154784,
|
| 1987 |
+
"eval_biology_loss": 3.621983528137207,
|
| 1988 |
+
"eval_biology_mean_token_accuracy": 0.4429086031913757,
|
| 1989 |
+
"eval_biology_num_tokens": 39854873.0,
|
| 1990 |
+
"eval_biology_runtime": 50.3472,
|
| 1991 |
+
"eval_biology_samples_per_second": 9.931,
|
| 1992 |
+
"eval_biology_steps_per_second": 2.483,
|
| 1993 |
+
"step": 1400
|
| 1994 |
+
},
|
| 1995 |
+
{
|
| 1996 |
+
"epoch": 2.24,
|
| 1997 |
+
"eval_chemistry_entropy": 1.7147236638069152,
|
| 1998 |
+
"eval_chemistry_loss": 1.8550182580947876,
|
| 1999 |
+
"eval_chemistry_mean_token_accuracy": 0.6335213797092438,
|
| 2000 |
+
"eval_chemistry_num_tokens": 39854873.0,
|
| 2001 |
+
"eval_chemistry_runtime": 61.9675,
|
| 2002 |
+
"eval_chemistry_samples_per_second": 8.069,
|
| 2003 |
+
"eval_chemistry_steps_per_second": 2.017,
|
| 2004 |
+
"step": 1400
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 2.24,
|
| 2008 |
+
"eval_math_entropy": 0.7918919861316681,
|
| 2009 |
+
"eval_math_loss": 0.8215861320495605,
|
| 2010 |
+
"eval_math_mean_token_accuracy": 0.7919570498466492,
|
| 2011 |
+
"eval_math_num_tokens": 39854873.0,
|
| 2012 |
+
"eval_math_runtime": 63.4845,
|
| 2013 |
+
"eval_math_samples_per_second": 7.876,
|
| 2014 |
+
"eval_math_steps_per_second": 1.969,
|
| 2015 |
+
"step": 1400
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 2.24,
|
| 2019 |
+
"eval_physics_entropy": 1.364860999584198,
|
| 2020 |
+
"eval_physics_loss": 1.4583380222320557,
|
| 2021 |
+
"eval_physics_mean_token_accuracy": 0.6940155153274536,
|
| 2022 |
+
"eval_physics_num_tokens": 39854873.0,
|
| 2023 |
+
"eval_physics_runtime": 72.3589,
|
| 2024 |
+
"eval_physics_samples_per_second": 6.91,
|
| 2025 |
+
"eval_physics_steps_per_second": 1.728,
|
| 2026 |
+
"step": 1400
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"entropy": 0.7666558466851712,
|
| 2030 |
+
"epoch": 2.2560000000000002,
|
| 2031 |
+
"grad_norm": 16.875,
|
| 2032 |
+
"learning_rate": 1.1785185185185186e-05,
|
| 2033 |
+
"loss": 12.3089,
|
| 2034 |
+
"mean_token_accuracy": 0.8012977659702301,
|
| 2035 |
+
"num_tokens": 40141190.0,
|
| 2036 |
+
"step": 1410
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"entropy": 0.7879726745188236,
|
| 2040 |
+
"epoch": 2.2720000000000002,
|
| 2041 |
+
"grad_norm": 20.625,
|
| 2042 |
+
"learning_rate": 1.1711111111111113e-05,
|
| 2043 |
+
"loss": 12.6865,
|
| 2044 |
+
"mean_token_accuracy": 0.798141922056675,
|
| 2045 |
+
"num_tokens": 40415203.0,
|
| 2046 |
+
"step": 1420
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"entropy": 0.7712487172335386,
|
| 2050 |
+
"epoch": 2.288,
|
| 2051 |
+
"grad_norm": 16.875,
|
| 2052 |
+
"learning_rate": 1.1637037037037037e-05,
|
| 2053 |
+
"loss": 12.3102,
|
| 2054 |
+
"mean_token_accuracy": 0.800593600794673,
|
| 2055 |
+
"num_tokens": 40702393.0,
|
| 2056 |
+
"step": 1430
|
| 2057 |
+
},
|
| 2058 |
+
{
|
| 2059 |
+
"entropy": 0.7664103271439672,
|
| 2060 |
+
"epoch": 2.304,
|
| 2061 |
+
"grad_norm": 20.125,
|
| 2062 |
+
"learning_rate": 1.1562962962962964e-05,
|
| 2063 |
+
"loss": 12.3249,
|
| 2064 |
+
"mean_token_accuracy": 0.8032374054193496,
|
| 2065 |
+
"num_tokens": 40982775.0,
|
| 2066 |
+
"step": 1440
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"entropy": 0.7812536470592022,
|
| 2070 |
+
"epoch": 2.32,
|
| 2071 |
+
"grad_norm": 19.375,
|
| 2072 |
+
"learning_rate": 1.1488888888888889e-05,
|
| 2073 |
+
"loss": 12.5473,
|
| 2074 |
+
"mean_token_accuracy": 0.8004047337919473,
|
| 2075 |
+
"num_tokens": 41263356.0,
|
| 2076 |
+
"step": 1450
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"entropy": 0.7787353828549385,
|
| 2080 |
+
"epoch": 2.336,
|
| 2081 |
+
"grad_norm": 16.875,
|
| 2082 |
+
"learning_rate": 1.1414814814814817e-05,
|
| 2083 |
+
"loss": 12.5658,
|
| 2084 |
+
"mean_token_accuracy": 0.7996404368430376,
|
| 2085 |
+
"num_tokens": 41545235.0,
|
| 2086 |
+
"step": 1460
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"entropy": 0.7861184494569897,
|
| 2090 |
+
"epoch": 2.352,
|
| 2091 |
+
"grad_norm": 17.5,
|
| 2092 |
+
"learning_rate": 1.1340740740740742e-05,
|
| 2093 |
+
"loss": 12.6853,
|
| 2094 |
+
"mean_token_accuracy": 0.7966604385524988,
|
| 2095 |
+
"num_tokens": 41833417.0,
|
| 2096 |
+
"step": 1470
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"entropy": 0.772636204212904,
|
| 2100 |
+
"epoch": 2.368,
|
| 2101 |
+
"grad_norm": 18.5,
|
| 2102 |
+
"learning_rate": 1.1266666666666668e-05,
|
| 2103 |
+
"loss": 12.415,
|
| 2104 |
+
"mean_token_accuracy": 0.8015665214508771,
|
| 2105 |
+
"num_tokens": 42117030.0,
|
| 2106 |
+
"step": 1480
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"entropy": 0.7902130583301187,
|
| 2110 |
+
"epoch": 2.384,
|
| 2111 |
+
"grad_norm": 20.375,
|
| 2112 |
+
"learning_rate": 1.1192592592592593e-05,
|
| 2113 |
+
"loss": 12.7256,
|
| 2114 |
+
"mean_token_accuracy": 0.7966745045036078,
|
| 2115 |
+
"num_tokens": 42410990.0,
|
| 2116 |
+
"step": 1490
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"entropy": 0.7819203210994601,
|
| 2120 |
+
"epoch": 2.4,
|
| 2121 |
+
"grad_norm": 18.5,
|
| 2122 |
+
"learning_rate": 1.111851851851852e-05,
|
| 2123 |
+
"loss": 12.5447,
|
| 2124 |
+
"mean_token_accuracy": 0.8000232726335526,
|
| 2125 |
+
"num_tokens": 42691890.0,
|
| 2126 |
+
"step": 1500
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 2.4,
|
| 2130 |
+
"eval_biology_entropy": 3.1204140815734864,
|
| 2131 |
+
"eval_biology_loss": 3.581092357635498,
|
| 2132 |
+
"eval_biology_mean_token_accuracy": 0.4439757652282715,
|
| 2133 |
+
"eval_biology_num_tokens": 42691890.0,
|
| 2134 |
+
"eval_biology_runtime": 50.3432,
|
| 2135 |
+
"eval_biology_samples_per_second": 9.932,
|
| 2136 |
+
"eval_biology_steps_per_second": 2.483,
|
| 2137 |
+
"step": 1500
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.4,
|
| 2141 |
+
"eval_chemistry_entropy": 1.677762170791626,
|
| 2142 |
+
"eval_chemistry_loss": 1.8495266437530518,
|
| 2143 |
+
"eval_chemistry_mean_token_accuracy": 0.6339675810337067,
|
| 2144 |
+
"eval_chemistry_num_tokens": 42691890.0,
|
| 2145 |
+
"eval_chemistry_runtime": 61.8947,
|
| 2146 |
+
"eval_chemistry_samples_per_second": 8.078,
|
| 2147 |
+
"eval_chemistry_steps_per_second": 2.02,
|
| 2148 |
+
"step": 1500
|
| 2149 |
+
},
|
| 2150 |
+
{
|
| 2151 |
+
"epoch": 2.4,
|
| 2152 |
+
"eval_math_entropy": 0.7767732746601105,
|
| 2153 |
+
"eval_math_loss": 0.8190872073173523,
|
| 2154 |
+
"eval_math_mean_token_accuracy": 0.7922054343223571,
|
| 2155 |
+
"eval_math_num_tokens": 42691890.0,
|
| 2156 |
+
"eval_math_runtime": 63.4519,
|
| 2157 |
+
"eval_math_samples_per_second": 7.88,
|
| 2158 |
+
"eval_math_steps_per_second": 1.97,
|
| 2159 |
+
"step": 1500
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.4,
|
| 2163 |
+
"eval_physics_entropy": 1.3377611227035522,
|
| 2164 |
+
"eval_physics_loss": 1.4610776901245117,
|
| 2165 |
+
"eval_physics_mean_token_accuracy": 0.6939568905830383,
|
| 2166 |
+
"eval_physics_num_tokens": 42691890.0,
|
| 2167 |
+
"eval_physics_runtime": 72.3367,
|
| 2168 |
+
"eval_physics_samples_per_second": 6.912,
|
| 2169 |
+
"eval_physics_steps_per_second": 1.728,
|
| 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 sha256:9cb868f102d69fd1d8c1031ac30a1a24af52c3b42d8cb3460319a35dd624fb0f
|
| 3 |
+
size 6417
|