Upload folder using huggingface_hub
Browse files- .gitattributes +1 -0
- checkpoint-500/added_tokens.json +3 -0
- checkpoint-500/chat_template.jinja +47 -0
- checkpoint-500/config.json +110 -0
- checkpoint-500/generation_config.json +13 -0
- checkpoint-500/model-00001-of-00005.safetensors +3 -0
- checkpoint-500/model-00002-of-00005.safetensors +3 -0
- checkpoint-500/model-00003-of-00005.safetensors +3 -0
- checkpoint-500/model-00004-of-00005.safetensors +3 -0
- checkpoint-500/model-00005-of-00005.safetensors +3 -0
- checkpoint-500/model.safetensors.index.json +0 -0
- checkpoint-500/optimizer.pt +3 -0
- checkpoint-500/rng_state.pth +3 -0
- checkpoint-500/scheduler.pt +3 -0
- checkpoint-500/special_tokens_map.json +33 -0
- checkpoint-500/tokenizer.json +3 -0
- checkpoint-500/tokenizer.model +3 -0
- checkpoint-500/tokenizer_config.json +0 -0
- checkpoint-500/trainer_state.json +754 -0
- checkpoint-500/training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-500/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image_soft_token>": 262144
|
| 3 |
+
}
|
checkpoint-500/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-500/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-500/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-500/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-500/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-500/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-500/model-00004-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
|
| 3 |
+
size 4931296656
|
checkpoint-500/model-00005-of-00005.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b59b029fc72394122a3b10056775960e2afd7e87bf3f29b2e01db9f5f607c198
|
| 3 |
+
size 4601000928
|
checkpoint-500/model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3fc971f9b23c0c9812645587bcf5f4de40eba90fe2c598ecce008c52c24058f
|
| 3 |
+
size 896607944
|
checkpoint-500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fde986f97ecc270a34049538e5bcb495edbc4e7f5914a76b3cad083de287dc73
|
| 3 |
+
size 14645
|
checkpoint-500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ddc00081d4f892d131a0d2b4d4827a214275362bfa62dfb8413f489c1bfad457
|
| 3 |
+
size 1465
|
checkpoint-500/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-500/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
|
| 3 |
+
size 33384568
|
checkpoint-500/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
|
| 3 |
+
size 4689074
|
checkpoint-500/tokenizer_config.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-500/trainer_state.json
ADDED
|
@@ -0,0 +1,754 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.8901746967842439,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"entropy": 1.0559891855344177,
|
| 14 |
+
"epoch": 0.017803493935684877,
|
| 15 |
+
"grad_norm": 176.0,
|
| 16 |
+
"learning_rate": 6.000000000000001e-07,
|
| 17 |
+
"loss": 42.5063,
|
| 18 |
+
"mean_token_accuracy": 0.6153968568891287,
|
| 19 |
+
"num_tokens": 257189.0,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"entropy": 1.0444028429687022,
|
| 24 |
+
"epoch": 0.035606987871369754,
|
| 25 |
+
"grad_norm": 160.0,
|
| 26 |
+
"learning_rate": 1.2666666666666669e-06,
|
| 27 |
+
"loss": 41.7045,
|
| 28 |
+
"mean_token_accuracy": 0.619235553778708,
|
| 29 |
+
"num_tokens": 528285.0,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"entropy": 1.0654841650277376,
|
| 34 |
+
"epoch": 0.053410481807054634,
|
| 35 |
+
"grad_norm": 175.0,
|
| 36 |
+
"learning_rate": 1.9333333333333336e-06,
|
| 37 |
+
"loss": 41.559,
|
| 38 |
+
"mean_token_accuracy": 0.6179817667230963,
|
| 39 |
+
"num_tokens": 801635.0,
|
| 40 |
+
"step": 30
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"entropy": 1.0894893994554877,
|
| 44 |
+
"epoch": 0.07121397574273951,
|
| 45 |
+
"grad_norm": 154.0,
|
| 46 |
+
"learning_rate": 2.6e-06,
|
| 47 |
+
"loss": 41.6138,
|
| 48 |
+
"mean_token_accuracy": 0.6147882426157594,
|
| 49 |
+
"num_tokens": 1060812.0,
|
| 50 |
+
"step": 40
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"entropy": 1.0985747126862406,
|
| 54 |
+
"epoch": 0.08901746967842439,
|
| 55 |
+
"grad_norm": 141.0,
|
| 56 |
+
"learning_rate": 3.266666666666667e-06,
|
| 57 |
+
"loss": 39.4404,
|
| 58 |
+
"mean_token_accuracy": 0.6201398545876146,
|
| 59 |
+
"num_tokens": 1327380.0,
|
| 60 |
+
"step": 50
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"entropy": 1.162555592879653,
|
| 64 |
+
"epoch": 0.10682096361410927,
|
| 65 |
+
"grad_norm": 114.0,
|
| 66 |
+
"learning_rate": 3.9333333333333335e-06,
|
| 67 |
+
"loss": 38.4348,
|
| 68 |
+
"mean_token_accuracy": 0.6222015436738729,
|
| 69 |
+
"num_tokens": 1597405.0,
|
| 70 |
+
"step": 60
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"entropy": 1.3072715956717729,
|
| 74 |
+
"epoch": 0.12462445754979415,
|
| 75 |
+
"grad_norm": 76.5,
|
| 76 |
+
"learning_rate": 4.600000000000001e-06,
|
| 77 |
+
"loss": 37.3098,
|
| 78 |
+
"mean_token_accuracy": 0.6205647233873606,
|
| 79 |
+
"num_tokens": 1859684.0,
|
| 80 |
+
"step": 70
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"entropy": 1.47952934615314,
|
| 84 |
+
"epoch": 0.14242795148547902,
|
| 85 |
+
"grad_norm": 54.0,
|
| 86 |
+
"learning_rate": 5.2666666666666665e-06,
|
| 87 |
+
"loss": 35.3269,
|
| 88 |
+
"mean_token_accuracy": 0.6284260725602507,
|
| 89 |
+
"num_tokens": 2118068.0,
|
| 90 |
+
"step": 80
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"entropy": 1.5287989236414432,
|
| 94 |
+
"epoch": 0.1602314454211639,
|
| 95 |
+
"grad_norm": 37.25,
|
| 96 |
+
"learning_rate": 5.933333333333335e-06,
|
| 97 |
+
"loss": 31.5676,
|
| 98 |
+
"mean_token_accuracy": 0.65506336633116,
|
| 99 |
+
"num_tokens": 2388824.0,
|
| 100 |
+
"step": 90
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"entropy": 1.630833999812603,
|
| 104 |
+
"epoch": 0.17803493935684878,
|
| 105 |
+
"grad_norm": 28.75,
|
| 106 |
+
"learning_rate": 6.600000000000001e-06,
|
| 107 |
+
"loss": 30.3566,
|
| 108 |
+
"mean_token_accuracy": 0.65591501891613,
|
| 109 |
+
"num_tokens": 2644330.0,
|
| 110 |
+
"step": 100
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.17803493935684878,
|
| 114 |
+
"eval_biology_entropy": 3.4087772026062013,
|
| 115 |
+
"eval_biology_loss": 3.8683884143829346,
|
| 116 |
+
"eval_biology_mean_token_accuracy": 0.4378368608951569,
|
| 117 |
+
"eval_biology_num_tokens": 2644330.0,
|
| 118 |
+
"eval_biology_runtime": 57.5948,
|
| 119 |
+
"eval_biology_samples_per_second": 8.681,
|
| 120 |
+
"eval_biology_steps_per_second": 2.17,
|
| 121 |
+
"step": 100
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17803493935684878,
|
| 125 |
+
"eval_chemistry_entropy": 1.644574547290802,
|
| 126 |
+
"eval_chemistry_loss": 1.8279423713684082,
|
| 127 |
+
"eval_chemistry_mean_token_accuracy": 0.663385133266449,
|
| 128 |
+
"eval_chemistry_num_tokens": 2644330.0,
|
| 129 |
+
"eval_chemistry_runtime": 72.0267,
|
| 130 |
+
"eval_chemistry_samples_per_second": 6.942,
|
| 131 |
+
"eval_chemistry_steps_per_second": 1.735,
|
| 132 |
+
"step": 100
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.17803493935684878,
|
| 136 |
+
"eval_math_entropy": 1.2650005059242249,
|
| 137 |
+
"eval_math_loss": 1.795242428779602,
|
| 138 |
+
"eval_math_mean_token_accuracy": 0.6832415590286255,
|
| 139 |
+
"eval_math_num_tokens": 2644330.0,
|
| 140 |
+
"eval_math_runtime": 75.2166,
|
| 141 |
+
"eval_math_samples_per_second": 6.647,
|
| 142 |
+
"eval_math_steps_per_second": 1.662,
|
| 143 |
+
"step": 100
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.17803493935684878,
|
| 147 |
+
"eval_physics_entropy": 1.6315128216743469,
|
| 148 |
+
"eval_physics_loss": 1.8273048400878906,
|
| 149 |
+
"eval_physics_mean_token_accuracy": 0.6732162787914276,
|
| 150 |
+
"eval_physics_num_tokens": 2644330.0,
|
| 151 |
+
"eval_physics_runtime": 85.3066,
|
| 152 |
+
"eval_physics_samples_per_second": 5.861,
|
| 153 |
+
"eval_physics_steps_per_second": 1.465,
|
| 154 |
+
"step": 100
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"entropy": 1.558345663547516,
|
| 158 |
+
"epoch": 0.19583843329253367,
|
| 159 |
+
"grad_norm": 24.375,
|
| 160 |
+
"learning_rate": 7.266666666666668e-06,
|
| 161 |
+
"loss": 28.2409,
|
| 162 |
+
"mean_token_accuracy": 0.6733386002480983,
|
| 163 |
+
"num_tokens": 2913700.0,
|
| 164 |
+
"step": 110
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"entropy": 1.5497455261647701,
|
| 168 |
+
"epoch": 0.21364192722821854,
|
| 169 |
+
"grad_norm": 20.375,
|
| 170 |
+
"learning_rate": 7.933333333333334e-06,
|
| 171 |
+
"loss": 27.4174,
|
| 172 |
+
"mean_token_accuracy": 0.678030077368021,
|
| 173 |
+
"num_tokens": 3185255.0,
|
| 174 |
+
"step": 120
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"entropy": 1.5162339597940444,
|
| 178 |
+
"epoch": 0.2314454211639034,
|
| 179 |
+
"grad_norm": 20.5,
|
| 180 |
+
"learning_rate": 8.6e-06,
|
| 181 |
+
"loss": 26.3563,
|
| 182 |
+
"mean_token_accuracy": 0.6866102360188961,
|
| 183 |
+
"num_tokens": 3454750.0,
|
| 184 |
+
"step": 130
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"entropy": 1.5057988293468951,
|
| 188 |
+
"epoch": 0.2492489150995883,
|
| 189 |
+
"grad_norm": 19.375,
|
| 190 |
+
"learning_rate": 9.266666666666667e-06,
|
| 191 |
+
"loss": 26.0435,
|
| 192 |
+
"mean_token_accuracy": 0.689361485093832,
|
| 193 |
+
"num_tokens": 3719113.0,
|
| 194 |
+
"step": 140
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"entropy": 1.544687245413661,
|
| 198 |
+
"epoch": 0.26705240903527316,
|
| 199 |
+
"grad_norm": 18.375,
|
| 200 |
+
"learning_rate": 9.933333333333334e-06,
|
| 201 |
+
"loss": 26.1449,
|
| 202 |
+
"mean_token_accuracy": 0.6857006324455142,
|
| 203 |
+
"num_tokens": 3990505.0,
|
| 204 |
+
"step": 150
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"entropy": 1.4617379672825337,
|
| 208 |
+
"epoch": 0.28485590297095803,
|
| 209 |
+
"grad_norm": 17.25,
|
| 210 |
+
"learning_rate": 1.0600000000000002e-05,
|
| 211 |
+
"loss": 24.9336,
|
| 212 |
+
"mean_token_accuracy": 0.6990214973688126,
|
| 213 |
+
"num_tokens": 4267403.0,
|
| 214 |
+
"step": 160
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"entropy": 1.45880494043231,
|
| 218 |
+
"epoch": 0.30265939690664295,
|
| 219 |
+
"grad_norm": 17.375,
|
| 220 |
+
"learning_rate": 1.1266666666666668e-05,
|
| 221 |
+
"loss": 24.5033,
|
| 222 |
+
"mean_token_accuracy": 0.7016687501221895,
|
| 223 |
+
"num_tokens": 4535458.0,
|
| 224 |
+
"step": 170
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"entropy": 1.435447308793664,
|
| 228 |
+
"epoch": 0.3204628908423278,
|
| 229 |
+
"grad_norm": 20.375,
|
| 230 |
+
"learning_rate": 1.1933333333333335e-05,
|
| 231 |
+
"loss": 24.3963,
|
| 232 |
+
"mean_token_accuracy": 0.7021712277084589,
|
| 233 |
+
"num_tokens": 4796815.0,
|
| 234 |
+
"step": 180
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"entropy": 1.4029231216758489,
|
| 238 |
+
"epoch": 0.3382663847780127,
|
| 239 |
+
"grad_norm": 16.875,
|
| 240 |
+
"learning_rate": 1.2600000000000001e-05,
|
| 241 |
+
"loss": 23.6953,
|
| 242 |
+
"mean_token_accuracy": 0.7080240704119205,
|
| 243 |
+
"num_tokens": 5066948.0,
|
| 244 |
+
"step": 190
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"entropy": 1.4468292627483605,
|
| 248 |
+
"epoch": 0.35606987871369755,
|
| 249 |
+
"grad_norm": 17.75,
|
| 250 |
+
"learning_rate": 1.3266666666666668e-05,
|
| 251 |
+
"loss": 24.4323,
|
| 252 |
+
"mean_token_accuracy": 0.698475543037057,
|
| 253 |
+
"num_tokens": 5324751.0,
|
| 254 |
+
"step": 200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.35606987871369755,
|
| 258 |
+
"eval_biology_entropy": 2.8836843852996825,
|
| 259 |
+
"eval_biology_loss": 3.357414722442627,
|
| 260 |
+
"eval_biology_mean_token_accuracy": 0.48108551049232484,
|
| 261 |
+
"eval_biology_num_tokens": 5324751.0,
|
| 262 |
+
"eval_biology_runtime": 57.5893,
|
| 263 |
+
"eval_biology_samples_per_second": 8.682,
|
| 264 |
+
"eval_biology_steps_per_second": 2.171,
|
| 265 |
+
"step": 200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.35606987871369755,
|
| 269 |
+
"eval_chemistry_entropy": 1.4061724228858947,
|
| 270 |
+
"eval_chemistry_loss": 1.478424072265625,
|
| 271 |
+
"eval_chemistry_mean_token_accuracy": 0.7063784518241882,
|
| 272 |
+
"eval_chemistry_num_tokens": 5324751.0,
|
| 273 |
+
"eval_chemistry_runtime": 72.1317,
|
| 274 |
+
"eval_chemistry_samples_per_second": 6.932,
|
| 275 |
+
"eval_chemistry_steps_per_second": 1.733,
|
| 276 |
+
"step": 200
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.35606987871369755,
|
| 280 |
+
"eval_math_entropy": 1.2177042779922485,
|
| 281 |
+
"eval_math_loss": 1.5454598665237427,
|
| 282 |
+
"eval_math_mean_token_accuracy": 0.7047145981788635,
|
| 283 |
+
"eval_math_num_tokens": 5324751.0,
|
| 284 |
+
"eval_math_runtime": 75.1207,
|
| 285 |
+
"eval_math_samples_per_second": 6.656,
|
| 286 |
+
"eval_math_steps_per_second": 1.664,
|
| 287 |
+
"step": 200
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.35606987871369755,
|
| 291 |
+
"eval_physics_entropy": 1.4176100668907166,
|
| 292 |
+
"eval_physics_loss": 1.5222268104553223,
|
| 293 |
+
"eval_physics_mean_token_accuracy": 0.707690957069397,
|
| 294 |
+
"eval_physics_num_tokens": 5324751.0,
|
| 295 |
+
"eval_physics_runtime": 85.2085,
|
| 296 |
+
"eval_physics_samples_per_second": 5.868,
|
| 297 |
+
"eval_physics_steps_per_second": 1.467,
|
| 298 |
+
"step": 200
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"entropy": 1.3558454412966967,
|
| 302 |
+
"epoch": 0.3738733726493824,
|
| 303 |
+
"grad_norm": 19.0,
|
| 304 |
+
"learning_rate": 1.3933333333333334e-05,
|
| 305 |
+
"loss": 22.7949,
|
| 306 |
+
"mean_token_accuracy": 0.7136942774057389,
|
| 307 |
+
"num_tokens": 5585508.0,
|
| 308 |
+
"step": 210
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"entropy": 1.383369480073452,
|
| 312 |
+
"epoch": 0.39167686658506734,
|
| 313 |
+
"grad_norm": 19.375,
|
| 314 |
+
"learning_rate": 1.46e-05,
|
| 315 |
+
"loss": 23.1375,
|
| 316 |
+
"mean_token_accuracy": 0.7102958835661412,
|
| 317 |
+
"num_tokens": 5848889.0,
|
| 318 |
+
"step": 220
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"entropy": 1.338986362144351,
|
| 322 |
+
"epoch": 0.4094803605207522,
|
| 323 |
+
"grad_norm": 17.25,
|
| 324 |
+
"learning_rate": 1.5266666666666667e-05,
|
| 325 |
+
"loss": 22.3403,
|
| 326 |
+
"mean_token_accuracy": 0.7182383798062801,
|
| 327 |
+
"num_tokens": 6114855.0,
|
| 328 |
+
"step": 230
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"entropy": 1.3609099809080363,
|
| 332 |
+
"epoch": 0.4272838544564371,
|
| 333 |
+
"grad_norm": 17.75,
|
| 334 |
+
"learning_rate": 1.5933333333333336e-05,
|
| 335 |
+
"loss": 22.8066,
|
| 336 |
+
"mean_token_accuracy": 0.7140973150730133,
|
| 337 |
+
"num_tokens": 6378152.0,
|
| 338 |
+
"step": 240
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"entropy": 1.3484373219311236,
|
| 342 |
+
"epoch": 0.44508734839212194,
|
| 343 |
+
"grad_norm": 18.5,
|
| 344 |
+
"learning_rate": 1.66e-05,
|
| 345 |
+
"loss": 22.591,
|
| 346 |
+
"mean_token_accuracy": 0.713333373889327,
|
| 347 |
+
"num_tokens": 6637273.0,
|
| 348 |
+
"step": 250
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"entropy": 1.315424071252346,
|
| 352 |
+
"epoch": 0.4628908423278068,
|
| 353 |
+
"grad_norm": 19.0,
|
| 354 |
+
"learning_rate": 1.726666666666667e-05,
|
| 355 |
+
"loss": 21.9951,
|
| 356 |
+
"mean_token_accuracy": 0.7204139836132526,
|
| 357 |
+
"num_tokens": 6896684.0,
|
| 358 |
+
"step": 260
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"entropy": 1.32754250690341,
|
| 362 |
+
"epoch": 0.48069433626349173,
|
| 363 |
+
"grad_norm": 17.25,
|
| 364 |
+
"learning_rate": 1.7933333333333333e-05,
|
| 365 |
+
"loss": 21.9557,
|
| 366 |
+
"mean_token_accuracy": 0.7203923668712378,
|
| 367 |
+
"num_tokens": 7166608.0,
|
| 368 |
+
"step": 270
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"entropy": 1.2635501787066459,
|
| 372 |
+
"epoch": 0.4984978301991766,
|
| 373 |
+
"grad_norm": 17.0,
|
| 374 |
+
"learning_rate": 1.86e-05,
|
| 375 |
+
"loss": 21.113,
|
| 376 |
+
"mean_token_accuracy": 0.7295621318742633,
|
| 377 |
+
"num_tokens": 7444923.0,
|
| 378 |
+
"step": 280
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"entropy": 1.3078652255237102,
|
| 382 |
+
"epoch": 0.5163013241348615,
|
| 383 |
+
"grad_norm": 17.25,
|
| 384 |
+
"learning_rate": 1.926666666666667e-05,
|
| 385 |
+
"loss": 21.7315,
|
| 386 |
+
"mean_token_accuracy": 0.7203882545232773,
|
| 387 |
+
"num_tokens": 7706502.0,
|
| 388 |
+
"step": 290
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"entropy": 1.3434829246252775,
|
| 392 |
+
"epoch": 0.5341048180705463,
|
| 393 |
+
"grad_norm": 20.25,
|
| 394 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 395 |
+
"loss": 22.236,
|
| 396 |
+
"mean_token_accuracy": 0.7177267197519541,
|
| 397 |
+
"num_tokens": 7969704.0,
|
| 398 |
+
"step": 300
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.5341048180705463,
|
| 402 |
+
"eval_biology_entropy": 2.7598990201950073,
|
| 403 |
+
"eval_biology_loss": 3.1821975708007812,
|
| 404 |
+
"eval_biology_mean_token_accuracy": 0.4959620425701141,
|
| 405 |
+
"eval_biology_num_tokens": 7969704.0,
|
| 406 |
+
"eval_biology_runtime": 57.9508,
|
| 407 |
+
"eval_biology_samples_per_second": 8.628,
|
| 408 |
+
"eval_biology_steps_per_second": 2.157,
|
| 409 |
+
"step": 300
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5341048180705463,
|
| 413 |
+
"eval_chemistry_entropy": 1.2813276634216308,
|
| 414 |
+
"eval_chemistry_loss": 1.3571581840515137,
|
| 415 |
+
"eval_chemistry_mean_token_accuracy": 0.7210862817764282,
|
| 416 |
+
"eval_chemistry_num_tokens": 7969704.0,
|
| 417 |
+
"eval_chemistry_runtime": 72.2035,
|
| 418 |
+
"eval_chemistry_samples_per_second": 6.925,
|
| 419 |
+
"eval_chemistry_steps_per_second": 1.731,
|
| 420 |
+
"step": 300
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.5341048180705463,
|
| 424 |
+
"eval_math_entropy": 1.165593782901764,
|
| 425 |
+
"eval_math_loss": 1.472881555557251,
|
| 426 |
+
"eval_math_mean_token_accuracy": 0.7125356969833374,
|
| 427 |
+
"eval_math_num_tokens": 7969704.0,
|
| 428 |
+
"eval_math_runtime": 75.1477,
|
| 429 |
+
"eval_math_samples_per_second": 6.654,
|
| 430 |
+
"eval_math_steps_per_second": 1.663,
|
| 431 |
+
"step": 300
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 0.5341048180705463,
|
| 435 |
+
"eval_physics_entropy": 1.3238337383270264,
|
| 436 |
+
"eval_physics_loss": 1.4213438034057617,
|
| 437 |
+
"eval_physics_mean_token_accuracy": 0.7196516485214234,
|
| 438 |
+
"eval_physics_num_tokens": 7969704.0,
|
| 439 |
+
"eval_physics_runtime": 85.3194,
|
| 440 |
+
"eval_physics_samples_per_second": 5.86,
|
| 441 |
+
"eval_physics_steps_per_second": 1.465,
|
| 442 |
+
"step": 300
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"entropy": 1.284867750480771,
|
| 446 |
+
"epoch": 0.5519083120062312,
|
| 447 |
+
"grad_norm": 21.625,
|
| 448 |
+
"learning_rate": 1.9933333333333334e-05,
|
| 449 |
+
"loss": 21.4794,
|
| 450 |
+
"mean_token_accuracy": 0.7222346868366003,
|
| 451 |
+
"num_tokens": 8242162.0,
|
| 452 |
+
"step": 310
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"entropy": 1.2864018987864256,
|
| 456 |
+
"epoch": 0.5697118059419161,
|
| 457 |
+
"grad_norm": 18.5,
|
| 458 |
+
"learning_rate": 1.985925925925926e-05,
|
| 459 |
+
"loss": 21.4262,
|
| 460 |
+
"mean_token_accuracy": 0.724996630847454,
|
| 461 |
+
"num_tokens": 8497852.0,
|
| 462 |
+
"step": 320
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"entropy": 1.2846032541245223,
|
| 466 |
+
"epoch": 0.587515299877601,
|
| 467 |
+
"grad_norm": 21.0,
|
| 468 |
+
"learning_rate": 1.9785185185185187e-05,
|
| 469 |
+
"loss": 21.5347,
|
| 470 |
+
"mean_token_accuracy": 0.7234938707202673,
|
| 471 |
+
"num_tokens": 8757753.0,
|
| 472 |
+
"step": 330
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"entropy": 1.2416468027979135,
|
| 476 |
+
"epoch": 0.6053187938132859,
|
| 477 |
+
"grad_norm": 15.75,
|
| 478 |
+
"learning_rate": 1.971111111111111e-05,
|
| 479 |
+
"loss": 20.5325,
|
| 480 |
+
"mean_token_accuracy": 0.733549839258194,
|
| 481 |
+
"num_tokens": 9024677.0,
|
| 482 |
+
"step": 340
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"entropy": 1.2626060862094164,
|
| 486 |
+
"epoch": 0.6231222877489707,
|
| 487 |
+
"grad_norm": 17.75,
|
| 488 |
+
"learning_rate": 1.963703703703704e-05,
|
| 489 |
+
"loss": 21.0676,
|
| 490 |
+
"mean_token_accuracy": 0.7265046216547489,
|
| 491 |
+
"num_tokens": 9291760.0,
|
| 492 |
+
"step": 350
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"entropy": 1.228025446087122,
|
| 496 |
+
"epoch": 0.6409257816846556,
|
| 497 |
+
"grad_norm": 19.875,
|
| 498 |
+
"learning_rate": 1.9562962962962964e-05,
|
| 499 |
+
"loss": 20.3954,
|
| 500 |
+
"mean_token_accuracy": 0.7328283190727234,
|
| 501 |
+
"num_tokens": 9561091.0,
|
| 502 |
+
"step": 360
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"entropy": 1.2296066496521234,
|
| 506 |
+
"epoch": 0.6587292756203404,
|
| 507 |
+
"grad_norm": 18.875,
|
| 508 |
+
"learning_rate": 1.948888888888889e-05,
|
| 509 |
+
"loss": 20.5407,
|
| 510 |
+
"mean_token_accuracy": 0.7325904417783022,
|
| 511 |
+
"num_tokens": 9827272.0,
|
| 512 |
+
"step": 370
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"entropy": 1.2079377524554729,
|
| 516 |
+
"epoch": 0.6765327695560254,
|
| 517 |
+
"grad_norm": 17.125,
|
| 518 |
+
"learning_rate": 1.9414814814814817e-05,
|
| 519 |
+
"loss": 20.0342,
|
| 520 |
+
"mean_token_accuracy": 0.7388566959649324,
|
| 521 |
+
"num_tokens": 10096065.0,
|
| 522 |
+
"step": 380
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"entropy": 1.2025926019996405,
|
| 526 |
+
"epoch": 0.6943362634917103,
|
| 527 |
+
"grad_norm": 18.125,
|
| 528 |
+
"learning_rate": 1.9340740740740743e-05,
|
| 529 |
+
"loss": 19.8907,
|
| 530 |
+
"mean_token_accuracy": 0.7389910601079464,
|
| 531 |
+
"num_tokens": 10364601.0,
|
| 532 |
+
"step": 390
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"entropy": 1.215231117978692,
|
| 536 |
+
"epoch": 0.7121397574273951,
|
| 537 |
+
"grad_norm": 18.5,
|
| 538 |
+
"learning_rate": 1.926666666666667e-05,
|
| 539 |
+
"loss": 20.2245,
|
| 540 |
+
"mean_token_accuracy": 0.7353602629154921,
|
| 541 |
+
"num_tokens": 10633325.0,
|
| 542 |
+
"step": 400
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.7121397574273951,
|
| 546 |
+
"eval_biology_entropy": 2.8210434341430664,
|
| 547 |
+
"eval_biology_loss": 3.116989850997925,
|
| 548 |
+
"eval_biology_mean_token_accuracy": 0.5034433958530427,
|
| 549 |
+
"eval_biology_num_tokens": 10633325.0,
|
| 550 |
+
"eval_biology_runtime": 57.5361,
|
| 551 |
+
"eval_biology_samples_per_second": 8.69,
|
| 552 |
+
"eval_biology_steps_per_second": 2.173,
|
| 553 |
+
"step": 400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.7121397574273951,
|
| 557 |
+
"eval_chemistry_entropy": 1.261210060596466,
|
| 558 |
+
"eval_chemistry_loss": 1.294067144393921,
|
| 559 |
+
"eval_chemistry_mean_token_accuracy": 0.730656672000885,
|
| 560 |
+
"eval_chemistry_num_tokens": 10633325.0,
|
| 561 |
+
"eval_chemistry_runtime": 72.1604,
|
| 562 |
+
"eval_chemistry_samples_per_second": 6.929,
|
| 563 |
+
"eval_chemistry_steps_per_second": 1.732,
|
| 564 |
+
"step": 400
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 0.7121397574273951,
|
| 568 |
+
"eval_math_entropy": 1.1643331799507142,
|
| 569 |
+
"eval_math_loss": 1.4408313035964966,
|
| 570 |
+
"eval_math_mean_token_accuracy": 0.7161806149482727,
|
| 571 |
+
"eval_math_num_tokens": 10633325.0,
|
| 572 |
+
"eval_math_runtime": 75.5006,
|
| 573 |
+
"eval_math_samples_per_second": 6.622,
|
| 574 |
+
"eval_math_steps_per_second": 1.656,
|
| 575 |
+
"step": 400
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 0.7121397574273951,
|
| 579 |
+
"eval_physics_entropy": 1.3186150641441345,
|
| 580 |
+
"eval_physics_loss": 1.373258113861084,
|
| 581 |
+
"eval_physics_mean_token_accuracy": 0.7264462962150574,
|
| 582 |
+
"eval_physics_num_tokens": 10633325.0,
|
| 583 |
+
"eval_physics_runtime": 85.0734,
|
| 584 |
+
"eval_physics_samples_per_second": 5.877,
|
| 585 |
+
"eval_physics_steps_per_second": 1.469,
|
| 586 |
+
"step": 400
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"entropy": 1.1908438373357058,
|
| 590 |
+
"epoch": 0.72994325136308,
|
| 591 |
+
"grad_norm": 22.875,
|
| 592 |
+
"learning_rate": 1.9192592592592593e-05,
|
| 593 |
+
"loss": 19.969,
|
| 594 |
+
"mean_token_accuracy": 0.7395484477281571,
|
| 595 |
+
"num_tokens": 10897916.0,
|
| 596 |
+
"step": 410
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"entropy": 1.2167691864073276,
|
| 600 |
+
"epoch": 0.7477467452987648,
|
| 601 |
+
"grad_norm": 20.0,
|
| 602 |
+
"learning_rate": 1.911851851851852e-05,
|
| 603 |
+
"loss": 20.1452,
|
| 604 |
+
"mean_token_accuracy": 0.7378258787095546,
|
| 605 |
+
"num_tokens": 11165356.0,
|
| 606 |
+
"step": 420
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"entropy": 1.2289611756801606,
|
| 610 |
+
"epoch": 0.7655502392344498,
|
| 611 |
+
"grad_norm": 18.75,
|
| 612 |
+
"learning_rate": 1.9044444444444446e-05,
|
| 613 |
+
"loss": 20.5277,
|
| 614 |
+
"mean_token_accuracy": 0.7328679781407118,
|
| 615 |
+
"num_tokens": 11436799.0,
|
| 616 |
+
"step": 430
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"entropy": 1.244011626765132,
|
| 620 |
+
"epoch": 0.7833537331701347,
|
| 621 |
+
"grad_norm": 19.875,
|
| 622 |
+
"learning_rate": 1.8970370370370372e-05,
|
| 623 |
+
"loss": 20.5562,
|
| 624 |
+
"mean_token_accuracy": 0.7318485330790281,
|
| 625 |
+
"num_tokens": 11703496.0,
|
| 626 |
+
"step": 440
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"entropy": 1.2247176449745893,
|
| 630 |
+
"epoch": 0.8011572271058195,
|
| 631 |
+
"grad_norm": 21.75,
|
| 632 |
+
"learning_rate": 1.8896296296296295e-05,
|
| 633 |
+
"loss": 20.421,
|
| 634 |
+
"mean_token_accuracy": 0.7341447170823813,
|
| 635 |
+
"num_tokens": 11965530.0,
|
| 636 |
+
"step": 450
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"entropy": 1.2138673104345798,
|
| 640 |
+
"epoch": 0.8189607210415044,
|
| 641 |
+
"grad_norm": 17.375,
|
| 642 |
+
"learning_rate": 1.8822222222222225e-05,
|
| 643 |
+
"loss": 20.2052,
|
| 644 |
+
"mean_token_accuracy": 0.7365878012031317,
|
| 645 |
+
"num_tokens": 12224427.0,
|
| 646 |
+
"step": 460
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"entropy": 1.1654350489377976,
|
| 650 |
+
"epoch": 0.8367642149771892,
|
| 651 |
+
"grad_norm": 17.875,
|
| 652 |
+
"learning_rate": 1.874814814814815e-05,
|
| 653 |
+
"loss": 19.4104,
|
| 654 |
+
"mean_token_accuracy": 0.7445328518748283,
|
| 655 |
+
"num_tokens": 12509124.0,
|
| 656 |
+
"step": 470
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"entropy": 1.198500120639801,
|
| 660 |
+
"epoch": 0.8545677089128741,
|
| 661 |
+
"grad_norm": 16.75,
|
| 662 |
+
"learning_rate": 1.8674074074074075e-05,
|
| 663 |
+
"loss": 19.8297,
|
| 664 |
+
"mean_token_accuracy": 0.7386186074465513,
|
| 665 |
+
"num_tokens": 12778408.0,
|
| 666 |
+
"step": 480
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"entropy": 1.170426043868065,
|
| 670 |
+
"epoch": 0.8723712028485591,
|
| 671 |
+
"grad_norm": 18.5,
|
| 672 |
+
"learning_rate": 1.86e-05,
|
| 673 |
+
"loss": 19.4215,
|
| 674 |
+
"mean_token_accuracy": 0.743653716892004,
|
| 675 |
+
"num_tokens": 13046473.0,
|
| 676 |
+
"step": 490
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"entropy": 1.2106714222580195,
|
| 680 |
+
"epoch": 0.8901746967842439,
|
| 681 |
+
"grad_norm": 16.875,
|
| 682 |
+
"learning_rate": 1.8525925925925928e-05,
|
| 683 |
+
"loss": 20.1785,
|
| 684 |
+
"mean_token_accuracy": 0.7360058560967445,
|
| 685 |
+
"num_tokens": 13301659.0,
|
| 686 |
+
"step": 500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.8901746967842439,
|
| 690 |
+
"eval_biology_entropy": 2.77167928981781,
|
| 691 |
+
"eval_biology_loss": 3.050346612930298,
|
| 692 |
+
"eval_biology_mean_token_accuracy": 0.5097111368179321,
|
| 693 |
+
"eval_biology_num_tokens": 13301659.0,
|
| 694 |
+
"eval_biology_runtime": 58.0048,
|
| 695 |
+
"eval_biology_samples_per_second": 8.62,
|
| 696 |
+
"eval_biology_steps_per_second": 2.155,
|
| 697 |
+
"step": 500
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.8901746967842439,
|
| 701 |
+
"eval_chemistry_entropy": 1.2252461080551147,
|
| 702 |
+
"eval_chemistry_loss": 1.2589221000671387,
|
| 703 |
+
"eval_chemistry_mean_token_accuracy": 0.7352680149078369,
|
| 704 |
+
"eval_chemistry_num_tokens": 13301659.0,
|
| 705 |
+
"eval_chemistry_runtime": 71.4907,
|
| 706 |
+
"eval_chemistry_samples_per_second": 6.994,
|
| 707 |
+
"eval_chemistry_steps_per_second": 1.748,
|
| 708 |
+
"step": 500
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 0.8901746967842439,
|
| 712 |
+
"eval_math_entropy": 1.1505462565422058,
|
| 713 |
+
"eval_math_loss": 1.420082688331604,
|
| 714 |
+
"eval_math_mean_token_accuracy": 0.7192822680473328,
|
| 715 |
+
"eval_math_num_tokens": 13301659.0,
|
| 716 |
+
"eval_math_runtime": 74.8775,
|
| 717 |
+
"eval_math_samples_per_second": 6.678,
|
| 718 |
+
"eval_math_steps_per_second": 1.669,
|
| 719 |
+
"step": 500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 0.8901746967842439,
|
| 723 |
+
"eval_physics_entropy": 1.2864371166229247,
|
| 724 |
+
"eval_physics_loss": 1.344913125038147,
|
| 725 |
+
"eval_physics_mean_token_accuracy": 0.7303608517646789,
|
| 726 |
+
"eval_physics_num_tokens": 13301659.0,
|
| 727 |
+
"eval_physics_runtime": 84.8812,
|
| 728 |
+
"eval_physics_samples_per_second": 5.891,
|
| 729 |
+
"eval_physics_steps_per_second": 1.473,
|
| 730 |
+
"step": 500
|
| 731 |
+
}
|
| 732 |
+
],
|
| 733 |
+
"logging_steps": 10,
|
| 734 |
+
"max_steps": 3000,
|
| 735 |
+
"num_input_tokens_seen": 0,
|
| 736 |
+
"num_train_epochs": 6,
|
| 737 |
+
"save_steps": 500,
|
| 738 |
+
"stateful_callbacks": {
|
| 739 |
+
"TrainerControl": {
|
| 740 |
+
"args": {
|
| 741 |
+
"should_epoch_stop": false,
|
| 742 |
+
"should_evaluate": false,
|
| 743 |
+
"should_log": false,
|
| 744 |
+
"should_save": true,
|
| 745 |
+
"should_training_stop": false
|
| 746 |
+
},
|
| 747 |
+
"attributes": {}
|
| 748 |
+
}
|
| 749 |
+
},
|
| 750 |
+
"total_flos": 1.3874307456966482e+18,
|
| 751 |
+
"train_batch_size": 4,
|
| 752 |
+
"trial_name": null,
|
| 753 |
+
"trial_params": null
|
| 754 |
+
}
|
checkpoint-500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b9cfb65cc2fb9da22c96cfd5c2465b3fd02e2a43a4b3cef8022ea0a9cb1a776e
|
| 3 |
+
size 6417
|