Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- added_tokens.json +28 -0
- checkpoint-1350/added_tokens.json +28 -0
- checkpoint-1350/config.json +30 -0
- checkpoint-1350/generation_config.json +6 -0
- checkpoint-1350/merges.txt +0 -0
- checkpoint-1350/model.safetensors +3 -0
- checkpoint-1350/optimizer.pt +3 -0
- checkpoint-1350/rng_state.pth +3 -0
- checkpoint-1350/scaler.pt +3 -0
- checkpoint-1350/scheduler.pt +3 -0
- checkpoint-1350/special_tokens_map.json +25 -0
- checkpoint-1350/tokenizer.json +3 -0
- checkpoint-1350/tokenizer_config.json +240 -0
- checkpoint-1350/trainer_state.json +2059 -0
- checkpoint-1350/training_args.bin +3 -0
- checkpoint-1350/vocab.json +0 -0
- config.json +30 -0
- dpo_config.json +18 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- runs/Jun05_14-17-48_47dafa9566a9/events.out.tfevents.1749133094.47dafa9566a9.1112.0 +3 -0
- special_tokens_map.json +25 -0
- tokenizer.json +3 -0
- tokenizer_config.json +240 -0
- vocab.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ 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-1350/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
checkpoint-1350/added_tokens.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</think>": 151668,
|
| 3 |
+
"</tool_call>": 151658,
|
| 4 |
+
"</tool_response>": 151666,
|
| 5 |
+
"<think>": 151667,
|
| 6 |
+
"<tool_call>": 151657,
|
| 7 |
+
"<tool_response>": 151665,
|
| 8 |
+
"<|box_end|>": 151649,
|
| 9 |
+
"<|box_start|>": 151648,
|
| 10 |
+
"<|endoftext|>": 151643,
|
| 11 |
+
"<|file_sep|>": 151664,
|
| 12 |
+
"<|fim_middle|>": 151660,
|
| 13 |
+
"<|fim_pad|>": 151662,
|
| 14 |
+
"<|fim_prefix|>": 151659,
|
| 15 |
+
"<|fim_suffix|>": 151661,
|
| 16 |
+
"<|im_end|>": 151645,
|
| 17 |
+
"<|im_start|>": 151644,
|
| 18 |
+
"<|image_pad|>": 151655,
|
| 19 |
+
"<|object_ref_end|>": 151647,
|
| 20 |
+
"<|object_ref_start|>": 151646,
|
| 21 |
+
"<|quad_end|>": 151651,
|
| 22 |
+
"<|quad_start|>": 151650,
|
| 23 |
+
"<|repo_name|>": 151663,
|
| 24 |
+
"<|video_pad|>": 151656,
|
| 25 |
+
"<|vision_end|>": 151653,
|
| 26 |
+
"<|vision_pad|>": 151654,
|
| 27 |
+
"<|vision_start|>": 151652
|
| 28 |
+
}
|
checkpoint-1350/config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151643,
|
| 9 |
+
"head_dim": 128,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"max_position_embeddings": 32768,
|
| 15 |
+
"max_window_layers": 28,
|
| 16 |
+
"model_type": "qwen3",
|
| 17 |
+
"num_attention_heads": 16,
|
| 18 |
+
"num_hidden_layers": 28,
|
| 19 |
+
"num_key_value_heads": 8,
|
| 20 |
+
"rms_norm_eps": 1e-06,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rope_theta": 1000000,
|
| 23 |
+
"sliding_window": null,
|
| 24 |
+
"tie_word_embeddings": true,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.52.3",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"use_sliding_window": false,
|
| 29 |
+
"vocab_size": 151936
|
| 30 |
+
}
|
checkpoint-1350/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.52.3"
|
| 6 |
+
}
|
checkpoint-1350/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1350/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a793c47676550a51ebca3f46d38c9caf6f35ccebbc241c469e660d591a568bc0
|
| 3 |
+
size 2384234968
|
checkpoint-1350/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:337a963dbd5ae1571ab463f145ea25d41b3736353cde0dfc162c5a3ae3185370
|
| 3 |
+
size 4768662910
|
checkpoint-1350/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
|
| 3 |
+
size 14244
|
checkpoint-1350/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67b478be7422409187c8570cd62ff2d6c1f23d07dbe65a32454f0d210b437c34
|
| 3 |
+
size 988
|
checkpoint-1350/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bdd836cfabec29ee2868162191588ec230736f86630fa4f89c3fb14d0bf6fcc3
|
| 3 |
+
size 1064
|
checkpoint-1350/special_tokens_map.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"eos_token": {
|
| 18 |
+
"content": "<|endoftext|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>"
|
| 25 |
+
}
|
checkpoint-1350/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
|
| 3 |
+
size 11422654
|
checkpoint-1350/tokenizer_config.json
ADDED
|
@@ -0,0 +1,240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"151665": {
|
| 182 |
+
"content": "<tool_response>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"151666": {
|
| 190 |
+
"content": "</tool_response>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"151667": {
|
| 198 |
+
"content": "<think>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"151668": {
|
| 206 |
+
"content": "</think>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|im_start|>",
|
| 216 |
+
"<|im_end|>",
|
| 217 |
+
"<|object_ref_start|>",
|
| 218 |
+
"<|object_ref_end|>",
|
| 219 |
+
"<|box_start|>",
|
| 220 |
+
"<|box_end|>",
|
| 221 |
+
"<|quad_start|>",
|
| 222 |
+
"<|quad_end|>",
|
| 223 |
+
"<|vision_start|>",
|
| 224 |
+
"<|vision_end|>",
|
| 225 |
+
"<|vision_pad|>",
|
| 226 |
+
"<|image_pad|>",
|
| 227 |
+
"<|video_pad|>"
|
| 228 |
+
],
|
| 229 |
+
"bos_token": null,
|
| 230 |
+
"chat_template": null,
|
| 231 |
+
"clean_up_tokenization_spaces": false,
|
| 232 |
+
"eos_token": "<|endoftext|>",
|
| 233 |
+
"errors": "replace",
|
| 234 |
+
"extra_special_tokens": {},
|
| 235 |
+
"model_max_length": 131072,
|
| 236 |
+
"pad_token": "<|endoftext|>",
|
| 237 |
+
"split_special_tokens": false,
|
| 238 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 239 |
+
"unk_token": null
|
| 240 |
+
}
|
checkpoint-1350/trainer_state.json
ADDED
|
@@ -0,0 +1,2059 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1350,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.007410151908114116,
|
| 14 |
+
"grad_norm": 62.764404296875,
|
| 15 |
+
"learning_rate": 4.974074074074075e-06,
|
| 16 |
+
"logits/chosen": -1.1614253520965576,
|
| 17 |
+
"logits/rejected": -1.096572995185852,
|
| 18 |
+
"logps/chosen": -74.20314025878906,
|
| 19 |
+
"logps/rejected": -76.51347351074219,
|
| 20 |
+
"loss": 0.6586,
|
| 21 |
+
"rewards/accuracies": 0.5375000238418579,
|
| 22 |
+
"rewards/chosen": -0.20016133785247803,
|
| 23 |
+
"rewards/margins": 0.0967479944229126,
|
| 24 |
+
"rewards/rejected": -0.2969093322753906,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.014820303816228233,
|
| 29 |
+
"grad_norm": 79.51490020751953,
|
| 30 |
+
"learning_rate": 4.940740740740741e-06,
|
| 31 |
+
"logits/chosen": -1.5121562480926514,
|
| 32 |
+
"logits/rejected": -1.5445247888565063,
|
| 33 |
+
"logps/chosen": -85.74129486083984,
|
| 34 |
+
"logps/rejected": -93.8956527709961,
|
| 35 |
+
"loss": 0.6982,
|
| 36 |
+
"rewards/accuracies": 0.5,
|
| 37 |
+
"rewards/chosen": -1.1762564182281494,
|
| 38 |
+
"rewards/margins": 0.23413367569446564,
|
| 39 |
+
"rewards/rejected": -1.4103901386260986,
|
| 40 |
+
"step": 20
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.02223045572434235,
|
| 44 |
+
"grad_norm": 53.0880012512207,
|
| 45 |
+
"learning_rate": 4.907407407407408e-06,
|
| 46 |
+
"logits/chosen": -1.6576855182647705,
|
| 47 |
+
"logits/rejected": -1.526139259338379,
|
| 48 |
+
"logps/chosen": -86.98616027832031,
|
| 49 |
+
"logps/rejected": -96.59648132324219,
|
| 50 |
+
"loss": 0.5807,
|
| 51 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 52 |
+
"rewards/chosen": -1.2037205696105957,
|
| 53 |
+
"rewards/margins": 0.6045142412185669,
|
| 54 |
+
"rewards/rejected": -1.8082348108291626,
|
| 55 |
+
"step": 30
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.029640607632456465,
|
| 59 |
+
"grad_norm": 70.3161849975586,
|
| 60 |
+
"learning_rate": 4.870370370370371e-06,
|
| 61 |
+
"logits/chosen": -1.95455801486969,
|
| 62 |
+
"logits/rejected": -1.9040778875350952,
|
| 63 |
+
"logps/chosen": -86.21485900878906,
|
| 64 |
+
"logps/rejected": -90.55685424804688,
|
| 65 |
+
"loss": 0.7098,
|
| 66 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 67 |
+
"rewards/chosen": -1.4126113653182983,
|
| 68 |
+
"rewards/margins": 0.5408647656440735,
|
| 69 |
+
"rewards/rejected": -1.9534763097763062,
|
| 70 |
+
"step": 40
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.037050759540570584,
|
| 74 |
+
"grad_norm": 86.17560577392578,
|
| 75 |
+
"learning_rate": 4.833333333333333e-06,
|
| 76 |
+
"logits/chosen": -1.7684093713760376,
|
| 77 |
+
"logits/rejected": -1.6274621486663818,
|
| 78 |
+
"logps/chosen": -97.04888916015625,
|
| 79 |
+
"logps/rejected": -95.69999694824219,
|
| 80 |
+
"loss": 0.6707,
|
| 81 |
+
"rewards/accuracies": 0.5625,
|
| 82 |
+
"rewards/chosen": -1.532711386680603,
|
| 83 |
+
"rewards/margins": 0.37558549642562866,
|
| 84 |
+
"rewards/rejected": -1.9082969427108765,
|
| 85 |
+
"step": 50
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.0444609114486847,
|
| 89 |
+
"grad_norm": 64.5966796875,
|
| 90 |
+
"learning_rate": 4.800000000000001e-06,
|
| 91 |
+
"logits/chosen": -1.7628320455551147,
|
| 92 |
+
"logits/rejected": -1.7703828811645508,
|
| 93 |
+
"logps/chosen": -90.20537567138672,
|
| 94 |
+
"logps/rejected": -97.63748168945312,
|
| 95 |
+
"loss": 0.6683,
|
| 96 |
+
"rewards/accuracies": 0.5625,
|
| 97 |
+
"rewards/chosen": -1.3552738428115845,
|
| 98 |
+
"rewards/margins": 0.38691654801368713,
|
| 99 |
+
"rewards/rejected": -1.7421903610229492,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.051871063356798815,
|
| 104 |
+
"grad_norm": 85.56486511230469,
|
| 105 |
+
"learning_rate": 4.762962962962963e-06,
|
| 106 |
+
"logits/chosen": -1.415287733078003,
|
| 107 |
+
"logits/rejected": -1.4015685319900513,
|
| 108 |
+
"logps/chosen": -91.52272033691406,
|
| 109 |
+
"logps/rejected": -100.97013092041016,
|
| 110 |
+
"loss": 0.6871,
|
| 111 |
+
"rewards/accuracies": 0.675000011920929,
|
| 112 |
+
"rewards/chosen": -1.1221778392791748,
|
| 113 |
+
"rewards/margins": 0.47898730635643005,
|
| 114 |
+
"rewards/rejected": -1.6011650562286377,
|
| 115 |
+
"step": 70
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.05928121526491293,
|
| 119 |
+
"grad_norm": 47.64088439941406,
|
| 120 |
+
"learning_rate": 4.725925925925926e-06,
|
| 121 |
+
"logits/chosen": -1.169304609298706,
|
| 122 |
+
"logits/rejected": -1.170772910118103,
|
| 123 |
+
"logps/chosen": -80.82547760009766,
|
| 124 |
+
"logps/rejected": -93.60406494140625,
|
| 125 |
+
"loss": 0.5665,
|
| 126 |
+
"rewards/accuracies": 0.699999988079071,
|
| 127 |
+
"rewards/chosen": -0.8211374282836914,
|
| 128 |
+
"rewards/margins": 0.6154714822769165,
|
| 129 |
+
"rewards/rejected": -1.4366087913513184,
|
| 130 |
+
"step": 80
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.06669136717302705,
|
| 134 |
+
"grad_norm": 56.08961486816406,
|
| 135 |
+
"learning_rate": 4.6888888888888895e-06,
|
| 136 |
+
"logits/chosen": -1.2894606590270996,
|
| 137 |
+
"logits/rejected": -1.1954705715179443,
|
| 138 |
+
"logps/chosen": -84.70814514160156,
|
| 139 |
+
"logps/rejected": -93.02404022216797,
|
| 140 |
+
"loss": 0.6926,
|
| 141 |
+
"rewards/accuracies": 0.5625,
|
| 142 |
+
"rewards/chosen": -1.0803619623184204,
|
| 143 |
+
"rewards/margins": 0.41851162910461426,
|
| 144 |
+
"rewards/rejected": -1.4988737106323242,
|
| 145 |
+
"step": 90
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.07410151908114117,
|
| 149 |
+
"grad_norm": 51.74102020263672,
|
| 150 |
+
"learning_rate": 4.651851851851853e-06,
|
| 151 |
+
"logits/chosen": -1.5028270483016968,
|
| 152 |
+
"logits/rejected": -1.3893922567367554,
|
| 153 |
+
"logps/chosen": -88.91434478759766,
|
| 154 |
+
"logps/rejected": -91.33464813232422,
|
| 155 |
+
"loss": 0.6143,
|
| 156 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 157 |
+
"rewards/chosen": -1.174202561378479,
|
| 158 |
+
"rewards/margins": 0.4613053798675537,
|
| 159 |
+
"rewards/rejected": -1.6355078220367432,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.08151167098925528,
|
| 164 |
+
"grad_norm": 50.91643524169922,
|
| 165 |
+
"learning_rate": 4.614814814814815e-06,
|
| 166 |
+
"logits/chosen": -1.6547952890396118,
|
| 167 |
+
"logits/rejected": -1.4582024812698364,
|
| 168 |
+
"logps/chosen": -90.47889709472656,
|
| 169 |
+
"logps/rejected": -104.38134765625,
|
| 170 |
+
"loss": 0.5684,
|
| 171 |
+
"rewards/accuracies": 0.675000011920929,
|
| 172 |
+
"rewards/chosen": -1.2481868267059326,
|
| 173 |
+
"rewards/margins": 0.8181821703910828,
|
| 174 |
+
"rewards/rejected": -2.066368818283081,
|
| 175 |
+
"step": 110
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.0889218228973694,
|
| 179 |
+
"grad_norm": 94.5538330078125,
|
| 180 |
+
"learning_rate": 4.5777777777777785e-06,
|
| 181 |
+
"logits/chosen": -1.5053731203079224,
|
| 182 |
+
"logits/rejected": -1.3366467952728271,
|
| 183 |
+
"logps/chosen": -88.6790542602539,
|
| 184 |
+
"logps/rejected": -104.670166015625,
|
| 185 |
+
"loss": 0.5921,
|
| 186 |
+
"rewards/accuracies": 0.6875,
|
| 187 |
+
"rewards/chosen": -1.5242931842803955,
|
| 188 |
+
"rewards/margins": 0.9062445759773254,
|
| 189 |
+
"rewards/rejected": -2.430537700653076,
|
| 190 |
+
"step": 120
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.09633197480548351,
|
| 194 |
+
"grad_norm": 33.83641815185547,
|
| 195 |
+
"learning_rate": 4.540740740740741e-06,
|
| 196 |
+
"logits/chosen": -1.409499168395996,
|
| 197 |
+
"logits/rejected": -1.2274638414382935,
|
| 198 |
+
"logps/chosen": -85.58585357666016,
|
| 199 |
+
"logps/rejected": -108.52848815917969,
|
| 200 |
+
"loss": 0.4394,
|
| 201 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 202 |
+
"rewards/chosen": -1.420691967010498,
|
| 203 |
+
"rewards/margins": 1.632450819015503,
|
| 204 |
+
"rewards/rejected": -3.053142786026001,
|
| 205 |
+
"step": 130
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10374212671359763,
|
| 209 |
+
"grad_norm": 220.59970092773438,
|
| 210 |
+
"learning_rate": 4.503703703703704e-06,
|
| 211 |
+
"logits/chosen": -1.4208014011383057,
|
| 212 |
+
"logits/rejected": -1.2115123271942139,
|
| 213 |
+
"logps/chosen": -98.5712890625,
|
| 214 |
+
"logps/rejected": -114.5568618774414,
|
| 215 |
+
"loss": 0.5026,
|
| 216 |
+
"rewards/accuracies": 0.6875,
|
| 217 |
+
"rewards/chosen": -2.161109209060669,
|
| 218 |
+
"rewards/margins": 1.499983787536621,
|
| 219 |
+
"rewards/rejected": -3.661093235015869,
|
| 220 |
+
"step": 140
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.11115227862171174,
|
| 224 |
+
"grad_norm": 53.83354949951172,
|
| 225 |
+
"learning_rate": 4.4666666666666665e-06,
|
| 226 |
+
"logits/chosen": -1.5498363971710205,
|
| 227 |
+
"logits/rejected": -1.485797643661499,
|
| 228 |
+
"logps/chosen": -97.19245147705078,
|
| 229 |
+
"logps/rejected": -110.35697174072266,
|
| 230 |
+
"loss": 0.5838,
|
| 231 |
+
"rewards/accuracies": 0.637499988079071,
|
| 232 |
+
"rewards/chosen": -1.9847490787506104,
|
| 233 |
+
"rewards/margins": 1.1178936958312988,
|
| 234 |
+
"rewards/rejected": -3.102642774581909,
|
| 235 |
+
"step": 150
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.11856243052982586,
|
| 239 |
+
"grad_norm": 89.90550231933594,
|
| 240 |
+
"learning_rate": 4.42962962962963e-06,
|
| 241 |
+
"logits/chosen": -1.5130503177642822,
|
| 242 |
+
"logits/rejected": -1.3890920877456665,
|
| 243 |
+
"logps/chosen": -93.79810333251953,
|
| 244 |
+
"logps/rejected": -108.19990539550781,
|
| 245 |
+
"loss": 0.6064,
|
| 246 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 247 |
+
"rewards/chosen": -1.9345242977142334,
|
| 248 |
+
"rewards/margins": 1.380746841430664,
|
| 249 |
+
"rewards/rejected": -3.3152713775634766,
|
| 250 |
+
"step": 160
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.12597258243794,
|
| 254 |
+
"grad_norm": 52.394100189208984,
|
| 255 |
+
"learning_rate": 4.392592592592593e-06,
|
| 256 |
+
"logits/chosen": -1.5120335817337036,
|
| 257 |
+
"logits/rejected": -1.4776674509048462,
|
| 258 |
+
"logps/chosen": -93.30475616455078,
|
| 259 |
+
"logps/rejected": -110.5815200805664,
|
| 260 |
+
"loss": 0.5675,
|
| 261 |
+
"rewards/accuracies": 0.637499988079071,
|
| 262 |
+
"rewards/chosen": -1.8305046558380127,
|
| 263 |
+
"rewards/margins": 1.1410784721374512,
|
| 264 |
+
"rewards/rejected": -2.9715828895568848,
|
| 265 |
+
"step": 170
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.1333827343460541,
|
| 269 |
+
"grad_norm": 55.45066833496094,
|
| 270 |
+
"learning_rate": 4.3555555555555555e-06,
|
| 271 |
+
"logits/chosen": -1.941173791885376,
|
| 272 |
+
"logits/rejected": -1.788526177406311,
|
| 273 |
+
"logps/chosen": -86.53892517089844,
|
| 274 |
+
"logps/rejected": -108.03669738769531,
|
| 275 |
+
"loss": 0.4828,
|
| 276 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 277 |
+
"rewards/chosen": -1.8137779235839844,
|
| 278 |
+
"rewards/margins": 1.234937310218811,
|
| 279 |
+
"rewards/rejected": -3.048715114593506,
|
| 280 |
+
"step": 180
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.14079288625416822,
|
| 284 |
+
"grad_norm": 50.05571746826172,
|
| 285 |
+
"learning_rate": 4.318518518518519e-06,
|
| 286 |
+
"logits/chosen": -1.8429396152496338,
|
| 287 |
+
"logits/rejected": -1.8329576253890991,
|
| 288 |
+
"logps/chosen": -94.84736633300781,
|
| 289 |
+
"logps/rejected": -114.12214660644531,
|
| 290 |
+
"loss": 0.5133,
|
| 291 |
+
"rewards/accuracies": 0.6875,
|
| 292 |
+
"rewards/chosen": -2.004016160964966,
|
| 293 |
+
"rewards/margins": 1.0727269649505615,
|
| 294 |
+
"rewards/rejected": -3.0767431259155273,
|
| 295 |
+
"step": 190
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.14820303816228234,
|
| 299 |
+
"grad_norm": 70.21109008789062,
|
| 300 |
+
"learning_rate": 4.281481481481482e-06,
|
| 301 |
+
"logits/chosen": -2.149946451187134,
|
| 302 |
+
"logits/rejected": -2.0653207302093506,
|
| 303 |
+
"logps/chosen": -96.63078308105469,
|
| 304 |
+
"logps/rejected": -110.8408432006836,
|
| 305 |
+
"loss": 0.5697,
|
| 306 |
+
"rewards/accuracies": 0.5874999761581421,
|
| 307 |
+
"rewards/chosen": -2.3063206672668457,
|
| 308 |
+
"rewards/margins": 0.9903339147567749,
|
| 309 |
+
"rewards/rejected": -3.296654462814331,
|
| 310 |
+
"step": 200
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.15561319007039645,
|
| 314 |
+
"grad_norm": 44.284324645996094,
|
| 315 |
+
"learning_rate": 4.244444444444445e-06,
|
| 316 |
+
"logits/chosen": -2.2816834449768066,
|
| 317 |
+
"logits/rejected": -2.0756924152374268,
|
| 318 |
+
"logps/chosen": -101.39938354492188,
|
| 319 |
+
"logps/rejected": -104.1975326538086,
|
| 320 |
+
"loss": 0.7565,
|
| 321 |
+
"rewards/accuracies": 0.574999988079071,
|
| 322 |
+
"rewards/chosen": -2.6701712608337402,
|
| 323 |
+
"rewards/margins": 0.539871096611023,
|
| 324 |
+
"rewards/rejected": -3.2100422382354736,
|
| 325 |
+
"step": 210
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.16302334197851057,
|
| 329 |
+
"grad_norm": 49.169559478759766,
|
| 330 |
+
"learning_rate": 4.207407407407408e-06,
|
| 331 |
+
"logits/chosen": -2.303546667098999,
|
| 332 |
+
"logits/rejected": -2.1084389686584473,
|
| 333 |
+
"logps/chosen": -93.49501037597656,
|
| 334 |
+
"logps/rejected": -111.30097961425781,
|
| 335 |
+
"loss": 0.5543,
|
| 336 |
+
"rewards/accuracies": 0.675000011920929,
|
| 337 |
+
"rewards/chosen": -1.9729340076446533,
|
| 338 |
+
"rewards/margins": 1.0684603452682495,
|
| 339 |
+
"rewards/rejected": -3.0413944721221924,
|
| 340 |
+
"step": 220
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.17043349388662468,
|
| 344 |
+
"grad_norm": 47.69071578979492,
|
| 345 |
+
"learning_rate": 4.170370370370371e-06,
|
| 346 |
+
"logits/chosen": -2.608030319213867,
|
| 347 |
+
"logits/rejected": -2.3824543952941895,
|
| 348 |
+
"logps/chosen": -89.42241668701172,
|
| 349 |
+
"logps/rejected": -109.88175964355469,
|
| 350 |
+
"loss": 0.5986,
|
| 351 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 352 |
+
"rewards/chosen": -1.9395774602890015,
|
| 353 |
+
"rewards/margins": 0.9270322918891907,
|
| 354 |
+
"rewards/rejected": -2.866609573364258,
|
| 355 |
+
"step": 230
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 0.1778436457947388,
|
| 359 |
+
"grad_norm": 51.891334533691406,
|
| 360 |
+
"learning_rate": 4.133333333333333e-06,
|
| 361 |
+
"logits/chosen": -2.294018268585205,
|
| 362 |
+
"logits/rejected": -2.186135768890381,
|
| 363 |
+
"logps/chosen": -91.37318420410156,
|
| 364 |
+
"logps/rejected": -103.55364990234375,
|
| 365 |
+
"loss": 0.5908,
|
| 366 |
+
"rewards/accuracies": 0.675000011920929,
|
| 367 |
+
"rewards/chosen": -1.9577686786651611,
|
| 368 |
+
"rewards/margins": 0.8284266591072083,
|
| 369 |
+
"rewards/rejected": -2.7861952781677246,
|
| 370 |
+
"step": 240
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.1852537977028529,
|
| 374 |
+
"grad_norm": 60.76057434082031,
|
| 375 |
+
"learning_rate": 4.0962962962962965e-06,
|
| 376 |
+
"logits/chosen": -2.3174843788146973,
|
| 377 |
+
"logits/rejected": -2.1223363876342773,
|
| 378 |
+
"logps/chosen": -94.4203109741211,
|
| 379 |
+
"logps/rejected": -108.79350280761719,
|
| 380 |
+
"loss": 0.5691,
|
| 381 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 382 |
+
"rewards/chosen": -1.9872305393218994,
|
| 383 |
+
"rewards/margins": 1.057706356048584,
|
| 384 |
+
"rewards/rejected": -3.0449366569519043,
|
| 385 |
+
"step": 250
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.19266394961096703,
|
| 389 |
+
"grad_norm": 63.53352737426758,
|
| 390 |
+
"learning_rate": 4.05925925925926e-06,
|
| 391 |
+
"logits/chosen": -2.0859274864196777,
|
| 392 |
+
"logits/rejected": -1.9589965343475342,
|
| 393 |
+
"logps/chosen": -94.96440124511719,
|
| 394 |
+
"logps/rejected": -102.25624084472656,
|
| 395 |
+
"loss": 0.7933,
|
| 396 |
+
"rewards/accuracies": 0.5625,
|
| 397 |
+
"rewards/chosen": -1.9980733394622803,
|
| 398 |
+
"rewards/margins": 0.4961894154548645,
|
| 399 |
+
"rewards/rejected": -2.4942626953125,
|
| 400 |
+
"step": 260
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.20007410151908114,
|
| 404 |
+
"grad_norm": 97.90867614746094,
|
| 405 |
+
"learning_rate": 4.022222222222222e-06,
|
| 406 |
+
"logits/chosen": -1.9325847625732422,
|
| 407 |
+
"logits/rejected": -1.7728859186172485,
|
| 408 |
+
"logps/chosen": -94.22940826416016,
|
| 409 |
+
"logps/rejected": -105.72850036621094,
|
| 410 |
+
"loss": 0.5711,
|
| 411 |
+
"rewards/accuracies": 0.699999988079071,
|
| 412 |
+
"rewards/chosen": -1.8600021600723267,
|
| 413 |
+
"rewards/margins": 0.9546739459037781,
|
| 414 |
+
"rewards/rejected": -2.81467604637146,
|
| 415 |
+
"step": 270
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.20748425342719526,
|
| 419 |
+
"grad_norm": 50.39332962036133,
|
| 420 |
+
"learning_rate": 3.9851851851851855e-06,
|
| 421 |
+
"logits/chosen": -1.6392498016357422,
|
| 422 |
+
"logits/rejected": -1.746097207069397,
|
| 423 |
+
"logps/chosen": -90.52095794677734,
|
| 424 |
+
"logps/rejected": -107.78398132324219,
|
| 425 |
+
"loss": 0.4994,
|
| 426 |
+
"rewards/accuracies": 0.699999988079071,
|
| 427 |
+
"rewards/chosen": -1.6299854516983032,
|
| 428 |
+
"rewards/margins": 1.126053810119629,
|
| 429 |
+
"rewards/rejected": -2.7560391426086426,
|
| 430 |
+
"step": 280
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.21489440533530937,
|
| 434 |
+
"grad_norm": 82.5980453491211,
|
| 435 |
+
"learning_rate": 3.948148148148149e-06,
|
| 436 |
+
"logits/chosen": -1.9976590871810913,
|
| 437 |
+
"logits/rejected": -1.591271996498108,
|
| 438 |
+
"logps/chosen": -97.7765121459961,
|
| 439 |
+
"logps/rejected": -106.0452880859375,
|
| 440 |
+
"loss": 0.5625,
|
| 441 |
+
"rewards/accuracies": 0.699999988079071,
|
| 442 |
+
"rewards/chosen": -1.6871846914291382,
|
| 443 |
+
"rewards/margins": 1.2763103246688843,
|
| 444 |
+
"rewards/rejected": -2.9634947776794434,
|
| 445 |
+
"step": 290
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 0.2223045572434235,
|
| 449 |
+
"grad_norm": 47.00569152832031,
|
| 450 |
+
"learning_rate": 3.911111111111112e-06,
|
| 451 |
+
"logits/chosen": -1.844430923461914,
|
| 452 |
+
"logits/rejected": -1.5602703094482422,
|
| 453 |
+
"logps/chosen": -95.04554748535156,
|
| 454 |
+
"logps/rejected": -110.90150451660156,
|
| 455 |
+
"loss": 0.5173,
|
| 456 |
+
"rewards/accuracies": 0.75,
|
| 457 |
+
"rewards/chosen": -1.7995688915252686,
|
| 458 |
+
"rewards/margins": 1.4488320350646973,
|
| 459 |
+
"rewards/rejected": -3.248400926589966,
|
| 460 |
+
"step": 300
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.2297147091515376,
|
| 464 |
+
"grad_norm": 80.46615600585938,
|
| 465 |
+
"learning_rate": 3.874074074074074e-06,
|
| 466 |
+
"logits/chosen": -1.9769401550292969,
|
| 467 |
+
"logits/rejected": -1.816585898399353,
|
| 468 |
+
"logps/chosen": -99.07120513916016,
|
| 469 |
+
"logps/rejected": -114.0129623413086,
|
| 470 |
+
"loss": 0.5575,
|
| 471 |
+
"rewards/accuracies": 0.6875,
|
| 472 |
+
"rewards/chosen": -2.19231915473938,
|
| 473 |
+
"rewards/margins": 1.2036749124526978,
|
| 474 |
+
"rewards/rejected": -3.395993709564209,
|
| 475 |
+
"step": 310
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 0.23712486105965172,
|
| 479 |
+
"grad_norm": 59.91643142700195,
|
| 480 |
+
"learning_rate": 3.837037037037038e-06,
|
| 481 |
+
"logits/chosen": -2.00789213180542,
|
| 482 |
+
"logits/rejected": -1.9294627904891968,
|
| 483 |
+
"logps/chosen": -100.81642150878906,
|
| 484 |
+
"logps/rejected": -114.23243713378906,
|
| 485 |
+
"loss": 0.6409,
|
| 486 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 487 |
+
"rewards/chosen": -2.405135154724121,
|
| 488 |
+
"rewards/margins": 0.944028377532959,
|
| 489 |
+
"rewards/rejected": -3.34916353225708,
|
| 490 |
+
"step": 320
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.24453501296776584,
|
| 494 |
+
"grad_norm": 71.39558410644531,
|
| 495 |
+
"learning_rate": 3.8000000000000005e-06,
|
| 496 |
+
"logits/chosen": -2.077759265899658,
|
| 497 |
+
"logits/rejected": -1.8107578754425049,
|
| 498 |
+
"logps/chosen": -99.47129821777344,
|
| 499 |
+
"logps/rejected": -104.8876953125,
|
| 500 |
+
"loss": 0.6597,
|
| 501 |
+
"rewards/accuracies": 0.6000000238418579,
|
| 502 |
+
"rewards/chosen": -2.003695011138916,
|
| 503 |
+
"rewards/margins": 0.9091768264770508,
|
| 504 |
+
"rewards/rejected": -2.912871837615967,
|
| 505 |
+
"step": 330
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.25194516487588,
|
| 509 |
+
"grad_norm": 37.14337921142578,
|
| 510 |
+
"learning_rate": 3.7629629629629633e-06,
|
| 511 |
+
"logits/chosen": -2.1400954723358154,
|
| 512 |
+
"logits/rejected": -2.0752546787261963,
|
| 513 |
+
"logps/chosen": -97.14881134033203,
|
| 514 |
+
"logps/rejected": -105.8427505493164,
|
| 515 |
+
"loss": 0.6113,
|
| 516 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 517 |
+
"rewards/chosen": -1.9141082763671875,
|
| 518 |
+
"rewards/margins": 0.7792128324508667,
|
| 519 |
+
"rewards/rejected": -2.6933212280273438,
|
| 520 |
+
"step": 340
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2593553167839941,
|
| 524 |
+
"grad_norm": 56.138423919677734,
|
| 525 |
+
"learning_rate": 3.725925925925926e-06,
|
| 526 |
+
"logits/chosen": -2.091094970703125,
|
| 527 |
+
"logits/rejected": -2.0105533599853516,
|
| 528 |
+
"logps/chosen": -92.09162139892578,
|
| 529 |
+
"logps/rejected": -110.2835922241211,
|
| 530 |
+
"loss": 0.492,
|
| 531 |
+
"rewards/accuracies": 0.737500011920929,
|
| 532 |
+
"rewards/chosen": -1.6331777572631836,
|
| 533 |
+
"rewards/margins": 1.2891775369644165,
|
| 534 |
+
"rewards/rejected": -2.9223551750183105,
|
| 535 |
+
"step": 350
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.2667654686921082,
|
| 539 |
+
"grad_norm": 52.79886245727539,
|
| 540 |
+
"learning_rate": 3.688888888888889e-06,
|
| 541 |
+
"logits/chosen": -2.2325026988983154,
|
| 542 |
+
"logits/rejected": -2.1436426639556885,
|
| 543 |
+
"logps/chosen": -89.82434844970703,
|
| 544 |
+
"logps/rejected": -112.33735656738281,
|
| 545 |
+
"loss": 0.4911,
|
| 546 |
+
"rewards/accuracies": 0.75,
|
| 547 |
+
"rewards/chosen": -1.5006159543991089,
|
| 548 |
+
"rewards/margins": 1.1897337436676025,
|
| 549 |
+
"rewards/rejected": -2.69035005569458,
|
| 550 |
+
"step": 360
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 0.2741756206002223,
|
| 554 |
+
"grad_norm": 150.0937042236328,
|
| 555 |
+
"learning_rate": 3.651851851851852e-06,
|
| 556 |
+
"logits/chosen": -1.8895237445831299,
|
| 557 |
+
"logits/rejected": -1.8900665044784546,
|
| 558 |
+
"logps/chosen": -103.93204498291016,
|
| 559 |
+
"logps/rejected": -112.97098541259766,
|
| 560 |
+
"loss": 0.5877,
|
| 561 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 562 |
+
"rewards/chosen": -2.1075615882873535,
|
| 563 |
+
"rewards/margins": 1.1100621223449707,
|
| 564 |
+
"rewards/rejected": -3.2176239490509033,
|
| 565 |
+
"step": 370
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 0.28158577250833644,
|
| 569 |
+
"grad_norm": 42.55241012573242,
|
| 570 |
+
"learning_rate": 3.614814814814815e-06,
|
| 571 |
+
"logits/chosen": -1.945433259010315,
|
| 572 |
+
"logits/rejected": -1.7756521701812744,
|
| 573 |
+
"logps/chosen": -106.47743225097656,
|
| 574 |
+
"logps/rejected": -114.155517578125,
|
| 575 |
+
"loss": 0.6501,
|
| 576 |
+
"rewards/accuracies": 0.737500011920929,
|
| 577 |
+
"rewards/chosen": -2.3606436252593994,
|
| 578 |
+
"rewards/margins": 1.0331517457962036,
|
| 579 |
+
"rewards/rejected": -3.3937950134277344,
|
| 580 |
+
"step": 380
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 0.28899592441645056,
|
| 584 |
+
"grad_norm": 52.49542236328125,
|
| 585 |
+
"learning_rate": 3.577777777777778e-06,
|
| 586 |
+
"logits/chosen": -1.8807388544082642,
|
| 587 |
+
"logits/rejected": -1.8252395391464233,
|
| 588 |
+
"logps/chosen": -102.06422424316406,
|
| 589 |
+
"logps/rejected": -112.83709716796875,
|
| 590 |
+
"loss": 0.6055,
|
| 591 |
+
"rewards/accuracies": 0.675000011920929,
|
| 592 |
+
"rewards/chosen": -2.23865008354187,
|
| 593 |
+
"rewards/margins": 0.9633029103279114,
|
| 594 |
+
"rewards/rejected": -3.201953172683716,
|
| 595 |
+
"step": 390
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.29640607632456467,
|
| 599 |
+
"grad_norm": 88.17601013183594,
|
| 600 |
+
"learning_rate": 3.540740740740741e-06,
|
| 601 |
+
"logits/chosen": -1.6389392614364624,
|
| 602 |
+
"logits/rejected": -1.5933473110198975,
|
| 603 |
+
"logps/chosen": -99.57685852050781,
|
| 604 |
+
"logps/rejected": -115.68553161621094,
|
| 605 |
+
"loss": 0.5417,
|
| 606 |
+
"rewards/accuracies": 0.675000011920929,
|
| 607 |
+
"rewards/chosen": -1.9246084690093994,
|
| 608 |
+
"rewards/margins": 1.4971764087677002,
|
| 609 |
+
"rewards/rejected": -3.4217846393585205,
|
| 610 |
+
"step": 400
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.3038162282326788,
|
| 614 |
+
"grad_norm": 62.808006286621094,
|
| 615 |
+
"learning_rate": 3.503703703703704e-06,
|
| 616 |
+
"logits/chosen": -1.5727903842926025,
|
| 617 |
+
"logits/rejected": -1.5283151865005493,
|
| 618 |
+
"logps/chosen": -93.6846694946289,
|
| 619 |
+
"logps/rejected": -106.69166564941406,
|
| 620 |
+
"loss": 0.5396,
|
| 621 |
+
"rewards/accuracies": 0.75,
|
| 622 |
+
"rewards/chosen": -2.084920883178711,
|
| 623 |
+
"rewards/margins": 1.2174980640411377,
|
| 624 |
+
"rewards/rejected": -3.3024184703826904,
|
| 625 |
+
"step": 410
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3112263801407929,
|
| 629 |
+
"grad_norm": 29.901016235351562,
|
| 630 |
+
"learning_rate": 3.4666666666666672e-06,
|
| 631 |
+
"logits/chosen": -1.862898826599121,
|
| 632 |
+
"logits/rejected": -1.6489702463150024,
|
| 633 |
+
"logps/chosen": -97.56269836425781,
|
| 634 |
+
"logps/rejected": -110.67304992675781,
|
| 635 |
+
"loss": 0.5257,
|
| 636 |
+
"rewards/accuracies": 0.737500011920929,
|
| 637 |
+
"rewards/chosen": -2.035705089569092,
|
| 638 |
+
"rewards/margins": 1.2200233936309814,
|
| 639 |
+
"rewards/rejected": -3.255728244781494,
|
| 640 |
+
"step": 420
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.318636532048907,
|
| 644 |
+
"grad_norm": 44.00906753540039,
|
| 645 |
+
"learning_rate": 3.42962962962963e-06,
|
| 646 |
+
"logits/chosen": -1.9901949167251587,
|
| 647 |
+
"logits/rejected": -1.7650057077407837,
|
| 648 |
+
"logps/chosen": -95.73258209228516,
|
| 649 |
+
"logps/rejected": -118.86299133300781,
|
| 650 |
+
"loss": 0.4368,
|
| 651 |
+
"rewards/accuracies": 0.75,
|
| 652 |
+
"rewards/chosen": -1.874515175819397,
|
| 653 |
+
"rewards/margins": 1.841843605041504,
|
| 654 |
+
"rewards/rejected": -3.7163589000701904,
|
| 655 |
+
"step": 430
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.32604668395702113,
|
| 659 |
+
"grad_norm": 124.45765686035156,
|
| 660 |
+
"learning_rate": 3.392592592592593e-06,
|
| 661 |
+
"logits/chosen": -1.9858062267303467,
|
| 662 |
+
"logits/rejected": -1.8073575496673584,
|
| 663 |
+
"logps/chosen": -106.0125732421875,
|
| 664 |
+
"logps/rejected": -122.56363677978516,
|
| 665 |
+
"loss": 0.5898,
|
| 666 |
+
"rewards/accuracies": 0.699999988079071,
|
| 667 |
+
"rewards/chosen": -2.860097646713257,
|
| 668 |
+
"rewards/margins": 1.184191346168518,
|
| 669 |
+
"rewards/rejected": -4.044289588928223,
|
| 670 |
+
"step": 440
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.33345683586513525,
|
| 674 |
+
"grad_norm": 156.9593963623047,
|
| 675 |
+
"learning_rate": 3.3555555555555557e-06,
|
| 676 |
+
"logits/chosen": -1.7876373529434204,
|
| 677 |
+
"logits/rejected": -1.6721996068954468,
|
| 678 |
+
"logps/chosen": -106.1432113647461,
|
| 679 |
+
"logps/rejected": -121.39786529541016,
|
| 680 |
+
"loss": 0.5335,
|
| 681 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 682 |
+
"rewards/chosen": -3.1309943199157715,
|
| 683 |
+
"rewards/margins": 1.0686304569244385,
|
| 684 |
+
"rewards/rejected": -4.199625015258789,
|
| 685 |
+
"step": 450
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 0.34086698777324936,
|
| 689 |
+
"grad_norm": 91.00648498535156,
|
| 690 |
+
"learning_rate": 3.3185185185185185e-06,
|
| 691 |
+
"logits/chosen": -2.0905487537384033,
|
| 692 |
+
"logits/rejected": -2.0243828296661377,
|
| 693 |
+
"logps/chosen": -99.6212387084961,
|
| 694 |
+
"logps/rejected": -112.38456726074219,
|
| 695 |
+
"loss": 0.4929,
|
| 696 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 697 |
+
"rewards/chosen": -2.6742300987243652,
|
| 698 |
+
"rewards/margins": 1.208742380142212,
|
| 699 |
+
"rewards/rejected": -3.882972240447998,
|
| 700 |
+
"step": 460
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 0.3482771396813635,
|
| 704 |
+
"grad_norm": 73.95698547363281,
|
| 705 |
+
"learning_rate": 3.281481481481482e-06,
|
| 706 |
+
"logits/chosen": -2.102916717529297,
|
| 707 |
+
"logits/rejected": -1.9817787408828735,
|
| 708 |
+
"logps/chosen": -107.4316177368164,
|
| 709 |
+
"logps/rejected": -120.3554916381836,
|
| 710 |
+
"loss": 0.607,
|
| 711 |
+
"rewards/accuracies": 0.637499988079071,
|
| 712 |
+
"rewards/chosen": -3.0798966884613037,
|
| 713 |
+
"rewards/margins": 1.1342341899871826,
|
| 714 |
+
"rewards/rejected": -4.2141313552856445,
|
| 715 |
+
"step": 470
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 0.3556872915894776,
|
| 719 |
+
"grad_norm": 115.34271240234375,
|
| 720 |
+
"learning_rate": 3.2444444444444446e-06,
|
| 721 |
+
"logits/chosen": -1.920265793800354,
|
| 722 |
+
"logits/rejected": -1.776262879371643,
|
| 723 |
+
"logps/chosen": -107.17350006103516,
|
| 724 |
+
"logps/rejected": -119.41255950927734,
|
| 725 |
+
"loss": 0.5724,
|
| 726 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 727 |
+
"rewards/chosen": -3.024296283721924,
|
| 728 |
+
"rewards/margins": 1.301172137260437,
|
| 729 |
+
"rewards/rejected": -4.325467586517334,
|
| 730 |
+
"step": 480
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3630974434975917,
|
| 734 |
+
"grad_norm": 124.22569274902344,
|
| 735 |
+
"learning_rate": 3.2074074074074075e-06,
|
| 736 |
+
"logits/chosen": -2.066603660583496,
|
| 737 |
+
"logits/rejected": -1.9234092235565186,
|
| 738 |
+
"logps/chosen": -107.19173431396484,
|
| 739 |
+
"logps/rejected": -125.5118637084961,
|
| 740 |
+
"loss": 0.5957,
|
| 741 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 742 |
+
"rewards/chosen": -3.1629674434661865,
|
| 743 |
+
"rewards/margins": 1.20828115940094,
|
| 744 |
+
"rewards/rejected": -4.371248722076416,
|
| 745 |
+
"step": 490
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.3705075954057058,
|
| 749 |
+
"grad_norm": 93.52369689941406,
|
| 750 |
+
"learning_rate": 3.1703703703703707e-06,
|
| 751 |
+
"logits/chosen": -2.049516201019287,
|
| 752 |
+
"logits/rejected": -1.80709707736969,
|
| 753 |
+
"logps/chosen": -108.04731750488281,
|
| 754 |
+
"logps/rejected": -131.86807250976562,
|
| 755 |
+
"loss": 0.4991,
|
| 756 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 757 |
+
"rewards/chosen": -2.871161699295044,
|
| 758 |
+
"rewards/margins": 1.387619972229004,
|
| 759 |
+
"rewards/rejected": -4.258781909942627,
|
| 760 |
+
"step": 500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.37791774731381994,
|
| 764 |
+
"grad_norm": 55.953941345214844,
|
| 765 |
+
"learning_rate": 3.133333333333334e-06,
|
| 766 |
+
"logits/chosen": -2.3617918491363525,
|
| 767 |
+
"logits/rejected": -2.0952062606811523,
|
| 768 |
+
"logps/chosen": -95.71025085449219,
|
| 769 |
+
"logps/rejected": -117.8790512084961,
|
| 770 |
+
"loss": 0.4238,
|
| 771 |
+
"rewards/accuracies": 0.7875000238418579,
|
| 772 |
+
"rewards/chosen": -2.4553141593933105,
|
| 773 |
+
"rewards/margins": 1.845973253250122,
|
| 774 |
+
"rewards/rejected": -4.301287651062012,
|
| 775 |
+
"step": 510
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 0.38532789922193406,
|
| 779 |
+
"grad_norm": 87.4388427734375,
|
| 780 |
+
"learning_rate": 3.096296296296297e-06,
|
| 781 |
+
"logits/chosen": -2.3576157093048096,
|
| 782 |
+
"logits/rejected": -2.3086090087890625,
|
| 783 |
+
"logps/chosen": -103.87492370605469,
|
| 784 |
+
"logps/rejected": -129.99880981445312,
|
| 785 |
+
"loss": 0.4485,
|
| 786 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 787 |
+
"rewards/chosen": -2.7980117797851562,
|
| 788 |
+
"rewards/margins": 1.933409333229065,
|
| 789 |
+
"rewards/rejected": -4.731420993804932,
|
| 790 |
+
"step": 520
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 0.39273805113004817,
|
| 794 |
+
"grad_norm": 139.21774291992188,
|
| 795 |
+
"learning_rate": 3.0592592592592596e-06,
|
| 796 |
+
"logits/chosen": -2.3830113410949707,
|
| 797 |
+
"logits/rejected": -2.236015796661377,
|
| 798 |
+
"logps/chosen": -105.96150970458984,
|
| 799 |
+
"logps/rejected": -123.8498306274414,
|
| 800 |
+
"loss": 0.4772,
|
| 801 |
+
"rewards/accuracies": 0.737500011920929,
|
| 802 |
+
"rewards/chosen": -2.7964818477630615,
|
| 803 |
+
"rewards/margins": 1.444065809249878,
|
| 804 |
+
"rewards/rejected": -4.2405476570129395,
|
| 805 |
+
"step": 530
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 0.4001482030381623,
|
| 809 |
+
"grad_norm": 60.24081802368164,
|
| 810 |
+
"learning_rate": 3.0222222222222225e-06,
|
| 811 |
+
"logits/chosen": -2.510646104812622,
|
| 812 |
+
"logits/rejected": -2.3236303329467773,
|
| 813 |
+
"logps/chosen": -106.36656188964844,
|
| 814 |
+
"logps/rejected": -126.50679016113281,
|
| 815 |
+
"loss": 0.5079,
|
| 816 |
+
"rewards/accuracies": 0.75,
|
| 817 |
+
"rewards/chosen": -3.144716739654541,
|
| 818 |
+
"rewards/margins": 1.5049692392349243,
|
| 819 |
+
"rewards/rejected": -4.649685859680176,
|
| 820 |
+
"step": 540
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.4075583549462764,
|
| 824 |
+
"grad_norm": 44.708290100097656,
|
| 825 |
+
"learning_rate": 2.9851851851851853e-06,
|
| 826 |
+
"logits/chosen": -2.6934173107147217,
|
| 827 |
+
"logits/rejected": -2.5574169158935547,
|
| 828 |
+
"logps/chosen": -106.81391906738281,
|
| 829 |
+
"logps/rejected": -124.56950378417969,
|
| 830 |
+
"loss": 0.546,
|
| 831 |
+
"rewards/accuracies": 0.737500011920929,
|
| 832 |
+
"rewards/chosen": -3.2360634803771973,
|
| 833 |
+
"rewards/margins": 1.5249487161636353,
|
| 834 |
+
"rewards/rejected": -4.761012077331543,
|
| 835 |
+
"step": 550
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.4149685068543905,
|
| 839 |
+
"grad_norm": 100.06108856201172,
|
| 840 |
+
"learning_rate": 2.948148148148148e-06,
|
| 841 |
+
"logits/chosen": -2.540220260620117,
|
| 842 |
+
"logits/rejected": -2.5645575523376465,
|
| 843 |
+
"logps/chosen": -105.29388427734375,
|
| 844 |
+
"logps/rejected": -127.27003479003906,
|
| 845 |
+
"loss": 0.4493,
|
| 846 |
+
"rewards/accuracies": 0.737500011920929,
|
| 847 |
+
"rewards/chosen": -3.06217885017395,
|
| 848 |
+
"rewards/margins": 1.8223583698272705,
|
| 849 |
+
"rewards/rejected": -4.884537696838379,
|
| 850 |
+
"step": 560
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.42237865876250463,
|
| 854 |
+
"grad_norm": 68.57003784179688,
|
| 855 |
+
"learning_rate": 2.9111111111111114e-06,
|
| 856 |
+
"logits/chosen": -2.6107475757598877,
|
| 857 |
+
"logits/rejected": -2.4408977031707764,
|
| 858 |
+
"logps/chosen": -102.4161605834961,
|
| 859 |
+
"logps/rejected": -123.5961685180664,
|
| 860 |
+
"loss": 0.4529,
|
| 861 |
+
"rewards/accuracies": 0.8125,
|
| 862 |
+
"rewards/chosen": -2.7161800861358643,
|
| 863 |
+
"rewards/margins": 1.7760255336761475,
|
| 864 |
+
"rewards/rejected": -4.492205619812012,
|
| 865 |
+
"step": 570
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.42978881067061875,
|
| 869 |
+
"grad_norm": 83.09320068359375,
|
| 870 |
+
"learning_rate": 2.874074074074074e-06,
|
| 871 |
+
"logits/chosen": -2.5648531913757324,
|
| 872 |
+
"logits/rejected": -2.4905104637145996,
|
| 873 |
+
"logps/chosen": -93.52774810791016,
|
| 874 |
+
"logps/rejected": -117.04872131347656,
|
| 875 |
+
"loss": 0.5508,
|
| 876 |
+
"rewards/accuracies": 0.75,
|
| 877 |
+
"rewards/chosen": -2.38288950920105,
|
| 878 |
+
"rewards/margins": 1.4900834560394287,
|
| 879 |
+
"rewards/rejected": -3.8729729652404785,
|
| 880 |
+
"step": 580
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.43719896257873286,
|
| 884 |
+
"grad_norm": 114.40435791015625,
|
| 885 |
+
"learning_rate": 2.837037037037037e-06,
|
| 886 |
+
"logits/chosen": -2.6075551509857178,
|
| 887 |
+
"logits/rejected": -2.3488199710845947,
|
| 888 |
+
"logps/chosen": -108.3016128540039,
|
| 889 |
+
"logps/rejected": -127.59623718261719,
|
| 890 |
+
"loss": 0.532,
|
| 891 |
+
"rewards/accuracies": 0.6499999761581421,
|
| 892 |
+
"rewards/chosen": -3.0478949546813965,
|
| 893 |
+
"rewards/margins": 1.5749890804290771,
|
| 894 |
+
"rewards/rejected": -4.6228837966918945,
|
| 895 |
+
"step": 590
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 0.444609114486847,
|
| 899 |
+
"grad_norm": 48.23006057739258,
|
| 900 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 901 |
+
"logits/chosen": -2.520981788635254,
|
| 902 |
+
"logits/rejected": -2.405494213104248,
|
| 903 |
+
"logps/chosen": -111.14179992675781,
|
| 904 |
+
"logps/rejected": -126.3864974975586,
|
| 905 |
+
"loss": 0.5698,
|
| 906 |
+
"rewards/accuracies": 0.75,
|
| 907 |
+
"rewards/chosen": -3.072885036468506,
|
| 908 |
+
"rewards/margins": 1.396822214126587,
|
| 909 |
+
"rewards/rejected": -4.469707012176514,
|
| 910 |
+
"step": 600
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 0.4520192663949611,
|
| 914 |
+
"grad_norm": 52.764652252197266,
|
| 915 |
+
"learning_rate": 2.7629629629629636e-06,
|
| 916 |
+
"logits/chosen": -2.569906711578369,
|
| 917 |
+
"logits/rejected": -2.4188952445983887,
|
| 918 |
+
"logps/chosen": -109.3153305053711,
|
| 919 |
+
"logps/rejected": -122.66573333740234,
|
| 920 |
+
"loss": 0.6618,
|
| 921 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 922 |
+
"rewards/chosen": -2.8832812309265137,
|
| 923 |
+
"rewards/margins": 1.2652729749679565,
|
| 924 |
+
"rewards/rejected": -4.14855432510376,
|
| 925 |
+
"step": 610
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.4594294183030752,
|
| 929 |
+
"grad_norm": 56.053504943847656,
|
| 930 |
+
"learning_rate": 2.7259259259259264e-06,
|
| 931 |
+
"logits/chosen": -2.3336873054504395,
|
| 932 |
+
"logits/rejected": -2.220808267593384,
|
| 933 |
+
"logps/chosen": -97.85954284667969,
|
| 934 |
+
"logps/rejected": -117.603759765625,
|
| 935 |
+
"loss": 0.5057,
|
| 936 |
+
"rewards/accuracies": 0.737500011920929,
|
| 937 |
+
"rewards/chosen": -2.430934190750122,
|
| 938 |
+
"rewards/margins": 1.6638364791870117,
|
| 939 |
+
"rewards/rejected": -4.094770908355713,
|
| 940 |
+
"step": 620
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4668395702111893,
|
| 944 |
+
"grad_norm": 55.09632110595703,
|
| 945 |
+
"learning_rate": 2.6888888888888892e-06,
|
| 946 |
+
"logits/chosen": -2.3458609580993652,
|
| 947 |
+
"logits/rejected": -2.090057611465454,
|
| 948 |
+
"logps/chosen": -99.01778411865234,
|
| 949 |
+
"logps/rejected": -116.4244155883789,
|
| 950 |
+
"loss": 0.5289,
|
| 951 |
+
"rewards/accuracies": 0.737500011920929,
|
| 952 |
+
"rewards/chosen": -2.8742129802703857,
|
| 953 |
+
"rewards/margins": 1.1919338703155518,
|
| 954 |
+
"rewards/rejected": -4.066147327423096,
|
| 955 |
+
"step": 630
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.47424972211930344,
|
| 959 |
+
"grad_norm": 46.486846923828125,
|
| 960 |
+
"learning_rate": 2.651851851851852e-06,
|
| 961 |
+
"logits/chosen": -2.6187617778778076,
|
| 962 |
+
"logits/rejected": -2.299057722091675,
|
| 963 |
+
"logps/chosen": -106.09327697753906,
|
| 964 |
+
"logps/rejected": -121.18121337890625,
|
| 965 |
+
"loss": 0.5003,
|
| 966 |
+
"rewards/accuracies": 0.737500011920929,
|
| 967 |
+
"rewards/chosen": -2.976048469543457,
|
| 968 |
+
"rewards/margins": 1.5380513668060303,
|
| 969 |
+
"rewards/rejected": -4.514100074768066,
|
| 970 |
+
"step": 640
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.48165987402741756,
|
| 974 |
+
"grad_norm": 66.68344116210938,
|
| 975 |
+
"learning_rate": 2.614814814814815e-06,
|
| 976 |
+
"logits/chosen": -2.4219138622283936,
|
| 977 |
+
"logits/rejected": -2.2749531269073486,
|
| 978 |
+
"logps/chosen": -101.74494934082031,
|
| 979 |
+
"logps/rejected": -117.6960678100586,
|
| 980 |
+
"loss": 0.548,
|
| 981 |
+
"rewards/accuracies": 0.6875,
|
| 982 |
+
"rewards/chosen": -3.033750534057617,
|
| 983 |
+
"rewards/margins": 1.507805347442627,
|
| 984 |
+
"rewards/rejected": -4.541555881500244,
|
| 985 |
+
"step": 650
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.48907002593553167,
|
| 989 |
+
"grad_norm": 63.8469123840332,
|
| 990 |
+
"learning_rate": 2.577777777777778e-06,
|
| 991 |
+
"logits/chosen": -2.4797866344451904,
|
| 992 |
+
"logits/rejected": -2.3341572284698486,
|
| 993 |
+
"logps/chosen": -106.57930755615234,
|
| 994 |
+
"logps/rejected": -124.18519592285156,
|
| 995 |
+
"loss": 0.5449,
|
| 996 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 997 |
+
"rewards/chosen": -3.2545254230499268,
|
| 998 |
+
"rewards/margins": 1.5712066888809204,
|
| 999 |
+
"rewards/rejected": -4.825732231140137,
|
| 1000 |
+
"step": 660
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.4964801778436458,
|
| 1004 |
+
"grad_norm": 36.812870025634766,
|
| 1005 |
+
"learning_rate": 2.540740740740741e-06,
|
| 1006 |
+
"logits/chosen": -2.4504082202911377,
|
| 1007 |
+
"logits/rejected": -2.317399501800537,
|
| 1008 |
+
"logps/chosen": -101.97000122070312,
|
| 1009 |
+
"logps/rejected": -122.93861389160156,
|
| 1010 |
+
"loss": 0.4462,
|
| 1011 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1012 |
+
"rewards/chosen": -2.8019375801086426,
|
| 1013 |
+
"rewards/margins": 1.7157955169677734,
|
| 1014 |
+
"rewards/rejected": -4.517733573913574,
|
| 1015 |
+
"step": 670
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 0.50389032975176,
|
| 1019 |
+
"grad_norm": 87.01313781738281,
|
| 1020 |
+
"learning_rate": 2.503703703703704e-06,
|
| 1021 |
+
"logits/chosen": -2.5619702339172363,
|
| 1022 |
+
"logits/rejected": -2.2939906120300293,
|
| 1023 |
+
"logps/chosen": -104.42756652832031,
|
| 1024 |
+
"logps/rejected": -122.24269104003906,
|
| 1025 |
+
"loss": 0.5498,
|
| 1026 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1027 |
+
"rewards/chosen": -2.752537250518799,
|
| 1028 |
+
"rewards/margins": 1.4413821697235107,
|
| 1029 |
+
"rewards/rejected": -4.1939191818237305,
|
| 1030 |
+
"step": 680
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.5113004816598741,
|
| 1034 |
+
"grad_norm": 33.50537872314453,
|
| 1035 |
+
"learning_rate": 2.466666666666667e-06,
|
| 1036 |
+
"logits/chosen": -2.397477626800537,
|
| 1037 |
+
"logits/rejected": -2.298727035522461,
|
| 1038 |
+
"logps/chosen": -101.79267883300781,
|
| 1039 |
+
"logps/rejected": -121.082763671875,
|
| 1040 |
+
"loss": 0.4498,
|
| 1041 |
+
"rewards/accuracies": 0.75,
|
| 1042 |
+
"rewards/chosen": -2.776843309402466,
|
| 1043 |
+
"rewards/margins": 1.5951542854309082,
|
| 1044 |
+
"rewards/rejected": -4.371997833251953,
|
| 1045 |
+
"step": 690
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5187106335679882,
|
| 1049 |
+
"grad_norm": 28.241764068603516,
|
| 1050 |
+
"learning_rate": 2.42962962962963e-06,
|
| 1051 |
+
"logits/chosen": -2.191267251968384,
|
| 1052 |
+
"logits/rejected": -2.1144583225250244,
|
| 1053 |
+
"logps/chosen": -105.1906967163086,
|
| 1054 |
+
"logps/rejected": -119.65342712402344,
|
| 1055 |
+
"loss": 0.3694,
|
| 1056 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 1057 |
+
"rewards/chosen": -2.89141583442688,
|
| 1058 |
+
"rewards/margins": 1.8759052753448486,
|
| 1059 |
+
"rewards/rejected": -4.7673211097717285,
|
| 1060 |
+
"step": 700
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.5261207854761023,
|
| 1064 |
+
"grad_norm": 93.26385498046875,
|
| 1065 |
+
"learning_rate": 2.3925925925925927e-06,
|
| 1066 |
+
"logits/chosen": -2.700456142425537,
|
| 1067 |
+
"logits/rejected": -2.5114896297454834,
|
| 1068 |
+
"logps/chosen": -102.400146484375,
|
| 1069 |
+
"logps/rejected": -129.38079833984375,
|
| 1070 |
+
"loss": 0.456,
|
| 1071 |
+
"rewards/accuracies": 0.737500011920929,
|
| 1072 |
+
"rewards/chosen": -2.760223627090454,
|
| 1073 |
+
"rewards/margins": 2.1241064071655273,
|
| 1074 |
+
"rewards/rejected": -4.8843302726745605,
|
| 1075 |
+
"step": 710
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 0.5335309373842164,
|
| 1079 |
+
"grad_norm": 43.060646057128906,
|
| 1080 |
+
"learning_rate": 2.3555555555555555e-06,
|
| 1081 |
+
"logits/chosen": -2.715698719024658,
|
| 1082 |
+
"logits/rejected": -2.641836643218994,
|
| 1083 |
+
"logps/chosen": -111.295166015625,
|
| 1084 |
+
"logps/rejected": -129.7608184814453,
|
| 1085 |
+
"loss": 0.5273,
|
| 1086 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1087 |
+
"rewards/chosen": -3.2993626594543457,
|
| 1088 |
+
"rewards/margins": 1.6410433053970337,
|
| 1089 |
+
"rewards/rejected": -4.940405368804932,
|
| 1090 |
+
"step": 720
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 0.5409410892923305,
|
| 1094 |
+
"grad_norm": 92.00672912597656,
|
| 1095 |
+
"learning_rate": 2.318518518518519e-06,
|
| 1096 |
+
"logits/chosen": -2.390772819519043,
|
| 1097 |
+
"logits/rejected": -2.2357537746429443,
|
| 1098 |
+
"logps/chosen": -106.68476867675781,
|
| 1099 |
+
"logps/rejected": -131.3654327392578,
|
| 1100 |
+
"loss": 0.5709,
|
| 1101 |
+
"rewards/accuracies": 0.699999988079071,
|
| 1102 |
+
"rewards/chosen": -3.0549259185791016,
|
| 1103 |
+
"rewards/margins": 1.6485790014266968,
|
| 1104 |
+
"rewards/rejected": -4.703505039215088,
|
| 1105 |
+
"step": 730
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.5483512412004447,
|
| 1109 |
+
"grad_norm": 57.26805114746094,
|
| 1110 |
+
"learning_rate": 2.2814814814814816e-06,
|
| 1111 |
+
"logits/chosen": -2.196866035461426,
|
| 1112 |
+
"logits/rejected": -2.0569887161254883,
|
| 1113 |
+
"logps/chosen": -100.42630004882812,
|
| 1114 |
+
"logps/rejected": -125.41304779052734,
|
| 1115 |
+
"loss": 0.4778,
|
| 1116 |
+
"rewards/accuracies": 0.75,
|
| 1117 |
+
"rewards/chosen": -2.8341736793518066,
|
| 1118 |
+
"rewards/margins": 1.9348560571670532,
|
| 1119 |
+
"rewards/rejected": -4.76902961730957,
|
| 1120 |
+
"step": 740
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.5557613931085588,
|
| 1124 |
+
"grad_norm": 58.62699508666992,
|
| 1125 |
+
"learning_rate": 2.2444444444444445e-06,
|
| 1126 |
+
"logits/chosen": -2.461791753768921,
|
| 1127 |
+
"logits/rejected": -2.336097240447998,
|
| 1128 |
+
"logps/chosen": -113.08296966552734,
|
| 1129 |
+
"logps/rejected": -132.7778778076172,
|
| 1130 |
+
"loss": 0.5236,
|
| 1131 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1132 |
+
"rewards/chosen": -3.270090103149414,
|
| 1133 |
+
"rewards/margins": 1.6124324798583984,
|
| 1134 |
+
"rewards/rejected": -4.882522106170654,
|
| 1135 |
+
"step": 750
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.5631715450166729,
|
| 1139 |
+
"grad_norm": 41.212913513183594,
|
| 1140 |
+
"learning_rate": 2.2074074074074077e-06,
|
| 1141 |
+
"logits/chosen": -2.4121994972229004,
|
| 1142 |
+
"logits/rejected": -2.247474193572998,
|
| 1143 |
+
"logps/chosen": -105.69578552246094,
|
| 1144 |
+
"logps/rejected": -122.80550384521484,
|
| 1145 |
+
"loss": 0.3995,
|
| 1146 |
+
"rewards/accuracies": 0.7875000238418579,
|
| 1147 |
+
"rewards/chosen": -2.9703726768493652,
|
| 1148 |
+
"rewards/margins": 1.6150137186050415,
|
| 1149 |
+
"rewards/rejected": -4.585386276245117,
|
| 1150 |
+
"step": 760
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.570581696924787,
|
| 1154 |
+
"grad_norm": 75.84191131591797,
|
| 1155 |
+
"learning_rate": 2.1703703703703705e-06,
|
| 1156 |
+
"logits/chosen": -2.296901226043701,
|
| 1157 |
+
"logits/rejected": -2.2925124168395996,
|
| 1158 |
+
"logps/chosen": -110.8564224243164,
|
| 1159 |
+
"logps/rejected": -127.59207916259766,
|
| 1160 |
+
"loss": 0.5596,
|
| 1161 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1162 |
+
"rewards/chosen": -3.3091530799865723,
|
| 1163 |
+
"rewards/margins": 1.2775094509124756,
|
| 1164 |
+
"rewards/rejected": -4.5866618156433105,
|
| 1165 |
+
"step": 770
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.5779918488329011,
|
| 1169 |
+
"grad_norm": 89.46887969970703,
|
| 1170 |
+
"learning_rate": 2.133333333333334e-06,
|
| 1171 |
+
"logits/chosen": -2.6353976726531982,
|
| 1172 |
+
"logits/rejected": -2.512312173843384,
|
| 1173 |
+
"logps/chosen": -102.060791015625,
|
| 1174 |
+
"logps/rejected": -132.6158447265625,
|
| 1175 |
+
"loss": 0.3622,
|
| 1176 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1177 |
+
"rewards/chosen": -2.7914113998413086,
|
| 1178 |
+
"rewards/margins": 2.160378932952881,
|
| 1179 |
+
"rewards/rejected": -4.951790809631348,
|
| 1180 |
+
"step": 780
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 0.5854020007410152,
|
| 1184 |
+
"grad_norm": 69.98067474365234,
|
| 1185 |
+
"learning_rate": 2.0962962962962966e-06,
|
| 1186 |
+
"logits/chosen": -2.48567533493042,
|
| 1187 |
+
"logits/rejected": -2.188729763031006,
|
| 1188 |
+
"logps/chosen": -97.02547454833984,
|
| 1189 |
+
"logps/rejected": -121.714111328125,
|
| 1190 |
+
"loss": 0.4915,
|
| 1191 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 1192 |
+
"rewards/chosen": -2.587202548980713,
|
| 1193 |
+
"rewards/margins": 1.8987194299697876,
|
| 1194 |
+
"rewards/rejected": -4.485922336578369,
|
| 1195 |
+
"step": 790
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 0.5928121526491293,
|
| 1199 |
+
"grad_norm": 22.259716033935547,
|
| 1200 |
+
"learning_rate": 2.0592592592592595e-06,
|
| 1201 |
+
"logits/chosen": -2.56459641456604,
|
| 1202 |
+
"logits/rejected": -2.396904945373535,
|
| 1203 |
+
"logps/chosen": -99.90885925292969,
|
| 1204 |
+
"logps/rejected": -127.66976165771484,
|
| 1205 |
+
"loss": 0.3965,
|
| 1206 |
+
"rewards/accuracies": 0.75,
|
| 1207 |
+
"rewards/chosen": -2.7819571495056152,
|
| 1208 |
+
"rewards/margins": 1.9999752044677734,
|
| 1209 |
+
"rewards/rejected": -4.781932830810547,
|
| 1210 |
+
"step": 800
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.6002223045572435,
|
| 1214 |
+
"grad_norm": 83.96212005615234,
|
| 1215 |
+
"learning_rate": 2.0222222222222223e-06,
|
| 1216 |
+
"logits/chosen": -2.694140672683716,
|
| 1217 |
+
"logits/rejected": -2.4873809814453125,
|
| 1218 |
+
"logps/chosen": -99.41667175292969,
|
| 1219 |
+
"logps/rejected": -127.27046203613281,
|
| 1220 |
+
"loss": 0.3876,
|
| 1221 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1222 |
+
"rewards/chosen": -2.891671657562256,
|
| 1223 |
+
"rewards/margins": 2.2859904766082764,
|
| 1224 |
+
"rewards/rejected": -5.177661895751953,
|
| 1225 |
+
"step": 810
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 0.6076324564653576,
|
| 1229 |
+
"grad_norm": 67.90824127197266,
|
| 1230 |
+
"learning_rate": 1.985185185185185e-06,
|
| 1231 |
+
"logits/chosen": -2.767648220062256,
|
| 1232 |
+
"logits/rejected": -2.585026264190674,
|
| 1233 |
+
"logps/chosen": -104.31745910644531,
|
| 1234 |
+
"logps/rejected": -125.73681640625,
|
| 1235 |
+
"loss": 0.5279,
|
| 1236 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1237 |
+
"rewards/chosen": -3.1707539558410645,
|
| 1238 |
+
"rewards/margins": 1.6347720623016357,
|
| 1239 |
+
"rewards/rejected": -4.8055267333984375,
|
| 1240 |
+
"step": 820
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.6150426083734717,
|
| 1244 |
+
"grad_norm": 33.568599700927734,
|
| 1245 |
+
"learning_rate": 1.9481481481481484e-06,
|
| 1246 |
+
"logits/chosen": -2.702396869659424,
|
| 1247 |
+
"logits/rejected": -2.4589433670043945,
|
| 1248 |
+
"logps/chosen": -103.00837707519531,
|
| 1249 |
+
"logps/rejected": -135.2923583984375,
|
| 1250 |
+
"loss": 0.3985,
|
| 1251 |
+
"rewards/accuracies": 0.824999988079071,
|
| 1252 |
+
"rewards/chosen": -3.202925443649292,
|
| 1253 |
+
"rewards/margins": 2.087952136993408,
|
| 1254 |
+
"rewards/rejected": -5.290877342224121,
|
| 1255 |
+
"step": 830
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6224527602815858,
|
| 1259 |
+
"grad_norm": 140.95526123046875,
|
| 1260 |
+
"learning_rate": 1.9111111111111112e-06,
|
| 1261 |
+
"logits/chosen": -2.784031629562378,
|
| 1262 |
+
"logits/rejected": -2.869075298309326,
|
| 1263 |
+
"logps/chosen": -113.4225845336914,
|
| 1264 |
+
"logps/rejected": -129.70352172851562,
|
| 1265 |
+
"loss": 0.5025,
|
| 1266 |
+
"rewards/accuracies": 0.737500011920929,
|
| 1267 |
+
"rewards/chosen": -3.5999884605407715,
|
| 1268 |
+
"rewards/margins": 1.6119463443756104,
|
| 1269 |
+
"rewards/rejected": -5.211935520172119,
|
| 1270 |
+
"step": 840
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6298629121896999,
|
| 1274 |
+
"grad_norm": 52.54026412963867,
|
| 1275 |
+
"learning_rate": 1.8740740740740743e-06,
|
| 1276 |
+
"logits/chosen": -2.8133158683776855,
|
| 1277 |
+
"logits/rejected": -2.6122548580169678,
|
| 1278 |
+
"logps/chosen": -111.7987289428711,
|
| 1279 |
+
"logps/rejected": -136.263671875,
|
| 1280 |
+
"loss": 0.4448,
|
| 1281 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1282 |
+
"rewards/chosen": -3.69873046875,
|
| 1283 |
+
"rewards/margins": 1.8727216720581055,
|
| 1284 |
+
"rewards/rejected": -5.5714521408081055,
|
| 1285 |
+
"step": 850
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.637273064097814,
|
| 1289 |
+
"grad_norm": 92.39305114746094,
|
| 1290 |
+
"learning_rate": 1.8407407407407409e-06,
|
| 1291 |
+
"logits/chosen": -2.6346824169158936,
|
| 1292 |
+
"logits/rejected": -2.5525918006896973,
|
| 1293 |
+
"logps/chosen": -126.9793930053711,
|
| 1294 |
+
"logps/rejected": -136.64266967773438,
|
| 1295 |
+
"loss": 0.6642,
|
| 1296 |
+
"rewards/accuracies": 0.6625000238418579,
|
| 1297 |
+
"rewards/chosen": -4.109932899475098,
|
| 1298 |
+
"rewards/margins": 1.4245363473892212,
|
| 1299 |
+
"rewards/rejected": -5.534468650817871,
|
| 1300 |
+
"step": 860
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 0.6446832160059282,
|
| 1304 |
+
"grad_norm": 123.8089370727539,
|
| 1305 |
+
"learning_rate": 1.803703703703704e-06,
|
| 1306 |
+
"logits/chosen": -2.7528748512268066,
|
| 1307 |
+
"logits/rejected": -2.5514931678771973,
|
| 1308 |
+
"logps/chosen": -108.72850036621094,
|
| 1309 |
+
"logps/rejected": -127.7916488647461,
|
| 1310 |
+
"loss": 0.5106,
|
| 1311 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1312 |
+
"rewards/chosen": -3.318143367767334,
|
| 1313 |
+
"rewards/margins": 2.0562808513641357,
|
| 1314 |
+
"rewards/rejected": -5.374424934387207,
|
| 1315 |
+
"step": 870
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 0.6520933679140423,
|
| 1319 |
+
"grad_norm": 42.69985580444336,
|
| 1320 |
+
"learning_rate": 1.7666666666666668e-06,
|
| 1321 |
+
"logits/chosen": -2.861846923828125,
|
| 1322 |
+
"logits/rejected": -2.5939736366271973,
|
| 1323 |
+
"logps/chosen": -106.8058090209961,
|
| 1324 |
+
"logps/rejected": -130.95132446289062,
|
| 1325 |
+
"loss": 0.3698,
|
| 1326 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 1327 |
+
"rewards/chosen": -3.020836353302002,
|
| 1328 |
+
"rewards/margins": 2.2084438800811768,
|
| 1329 |
+
"rewards/rejected": -5.2292799949646,
|
| 1330 |
+
"step": 880
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.6595035198221564,
|
| 1334 |
+
"grad_norm": 100.36827850341797,
|
| 1335 |
+
"learning_rate": 1.7296296296296298e-06,
|
| 1336 |
+
"logits/chosen": -2.8901448249816895,
|
| 1337 |
+
"logits/rejected": -2.638826608657837,
|
| 1338 |
+
"logps/chosen": -113.2393798828125,
|
| 1339 |
+
"logps/rejected": -136.57003784179688,
|
| 1340 |
+
"loss": 0.506,
|
| 1341 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1342 |
+
"rewards/chosen": -3.743712902069092,
|
| 1343 |
+
"rewards/margins": 1.8884308338165283,
|
| 1344 |
+
"rewards/rejected": -5.632143974304199,
|
| 1345 |
+
"step": 890
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.6669136717302705,
|
| 1349 |
+
"grad_norm": 47.63283157348633,
|
| 1350 |
+
"learning_rate": 1.6925925925925926e-06,
|
| 1351 |
+
"logits/chosen": -2.751713991165161,
|
| 1352 |
+
"logits/rejected": -2.6139540672302246,
|
| 1353 |
+
"logps/chosen": -117.19677734375,
|
| 1354 |
+
"logps/rejected": -131.9976806640625,
|
| 1355 |
+
"loss": 0.6344,
|
| 1356 |
+
"rewards/accuracies": 0.737500011920929,
|
| 1357 |
+
"rewards/chosen": -3.642165422439575,
|
| 1358 |
+
"rewards/margins": 1.7322568893432617,
|
| 1359 |
+
"rewards/rejected": -5.374422550201416,
|
| 1360 |
+
"step": 900
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6743238236383846,
|
| 1364 |
+
"grad_norm": 112.51361083984375,
|
| 1365 |
+
"learning_rate": 1.6555555555555559e-06,
|
| 1366 |
+
"logits/chosen": -2.636543035507202,
|
| 1367 |
+
"logits/rejected": -2.5468497276306152,
|
| 1368 |
+
"logps/chosen": -112.26319885253906,
|
| 1369 |
+
"logps/rejected": -135.6024932861328,
|
| 1370 |
+
"loss": 0.4688,
|
| 1371 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 1372 |
+
"rewards/chosen": -3.398538589477539,
|
| 1373 |
+
"rewards/margins": 1.8205482959747314,
|
| 1374 |
+
"rewards/rejected": -5.21908712387085,
|
| 1375 |
+
"step": 910
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.6817339755464987,
|
| 1379 |
+
"grad_norm": 63.83774185180664,
|
| 1380 |
+
"learning_rate": 1.6185185185185187e-06,
|
| 1381 |
+
"logits/chosen": -2.567595958709717,
|
| 1382 |
+
"logits/rejected": -2.4459292888641357,
|
| 1383 |
+
"logps/chosen": -113.2123031616211,
|
| 1384 |
+
"logps/rejected": -133.6845245361328,
|
| 1385 |
+
"loss": 0.5357,
|
| 1386 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1387 |
+
"rewards/chosen": -3.6509101390838623,
|
| 1388 |
+
"rewards/margins": 1.547790765762329,
|
| 1389 |
+
"rewards/rejected": -5.19870138168335,
|
| 1390 |
+
"step": 920
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 0.6891441274546128,
|
| 1394 |
+
"grad_norm": 97.52435302734375,
|
| 1395 |
+
"learning_rate": 1.5814814814814816e-06,
|
| 1396 |
+
"logits/chosen": -2.5026490688323975,
|
| 1397 |
+
"logits/rejected": -2.3349032402038574,
|
| 1398 |
+
"logps/chosen": -110.58065032958984,
|
| 1399 |
+
"logps/rejected": -134.26712036132812,
|
| 1400 |
+
"loss": 0.4725,
|
| 1401 |
+
"rewards/accuracies": 0.737500011920929,
|
| 1402 |
+
"rewards/chosen": -3.394016981124878,
|
| 1403 |
+
"rewards/margins": 1.8871221542358398,
|
| 1404 |
+
"rewards/rejected": -5.281139373779297,
|
| 1405 |
+
"step": 930
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 0.696554279362727,
|
| 1409 |
+
"grad_norm": 44.30360412597656,
|
| 1410 |
+
"learning_rate": 1.5444444444444446e-06,
|
| 1411 |
+
"logits/chosen": -2.691323757171631,
|
| 1412 |
+
"logits/rejected": -2.4929356575012207,
|
| 1413 |
+
"logps/chosen": -98.93566131591797,
|
| 1414 |
+
"logps/rejected": -121.6566162109375,
|
| 1415 |
+
"loss": 0.3602,
|
| 1416 |
+
"rewards/accuracies": 0.8374999761581421,
|
| 1417 |
+
"rewards/chosen": -2.962592601776123,
|
| 1418 |
+
"rewards/margins": 2.068859100341797,
|
| 1419 |
+
"rewards/rejected": -5.03145170211792,
|
| 1420 |
+
"step": 940
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 0.7039644312708411,
|
| 1424 |
+
"grad_norm": 38.775630950927734,
|
| 1425 |
+
"learning_rate": 1.5074074074074074e-06,
|
| 1426 |
+
"logits/chosen": -2.431483745574951,
|
| 1427 |
+
"logits/rejected": -2.3471839427948,
|
| 1428 |
+
"logps/chosen": -107.6336441040039,
|
| 1429 |
+
"logps/rejected": -129.42567443847656,
|
| 1430 |
+
"loss": 0.4332,
|
| 1431 |
+
"rewards/accuracies": 0.7875000238418579,
|
| 1432 |
+
"rewards/chosen": -3.1802752017974854,
|
| 1433 |
+
"rewards/margins": 1.8515145778656006,
|
| 1434 |
+
"rewards/rejected": -5.031789779663086,
|
| 1435 |
+
"step": 950
|
| 1436 |
+
},
|
| 1437 |
+
{
|
| 1438 |
+
"epoch": 0.7113745831789552,
|
| 1439 |
+
"grad_norm": 94.63509368896484,
|
| 1440 |
+
"learning_rate": 1.4703703703703707e-06,
|
| 1441 |
+
"logits/chosen": -2.6569230556488037,
|
| 1442 |
+
"logits/rejected": -2.5298376083374023,
|
| 1443 |
+
"logps/chosen": -112.77482604980469,
|
| 1444 |
+
"logps/rejected": -136.20745849609375,
|
| 1445 |
+
"loss": 0.5287,
|
| 1446 |
+
"rewards/accuracies": 0.737500011920929,
|
| 1447 |
+
"rewards/chosen": -3.546157121658325,
|
| 1448 |
+
"rewards/margins": 2.0040981769561768,
|
| 1449 |
+
"rewards/rejected": -5.55025577545166,
|
| 1450 |
+
"step": 960
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.7187847350870693,
|
| 1454 |
+
"grad_norm": 63.06848907470703,
|
| 1455 |
+
"learning_rate": 1.4333333333333335e-06,
|
| 1456 |
+
"logits/chosen": -2.6780967712402344,
|
| 1457 |
+
"logits/rejected": -2.4963746070861816,
|
| 1458 |
+
"logps/chosen": -102.239990234375,
|
| 1459 |
+
"logps/rejected": -125.75809478759766,
|
| 1460 |
+
"loss": 0.558,
|
| 1461 |
+
"rewards/accuracies": 0.6875,
|
| 1462 |
+
"rewards/chosen": -3.2254154682159424,
|
| 1463 |
+
"rewards/margins": 1.714345932006836,
|
| 1464 |
+
"rewards/rejected": -4.939761161804199,
|
| 1465 |
+
"step": 970
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7261948869951834,
|
| 1469 |
+
"grad_norm": 113.9366683959961,
|
| 1470 |
+
"learning_rate": 1.3962962962962963e-06,
|
| 1471 |
+
"logits/chosen": -2.5560450553894043,
|
| 1472 |
+
"logits/rejected": -2.4831178188323975,
|
| 1473 |
+
"logps/chosen": -114.14649963378906,
|
| 1474 |
+
"logps/rejected": -129.1222381591797,
|
| 1475 |
+
"loss": 0.6047,
|
| 1476 |
+
"rewards/accuracies": 0.6875,
|
| 1477 |
+
"rewards/chosen": -3.847761869430542,
|
| 1478 |
+
"rewards/margins": 1.4522716999053955,
|
| 1479 |
+
"rewards/rejected": -5.3000335693359375,
|
| 1480 |
+
"step": 980
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.7336050389032975,
|
| 1484 |
+
"grad_norm": 122.37875366210938,
|
| 1485 |
+
"learning_rate": 1.3592592592592594e-06,
|
| 1486 |
+
"logits/chosen": -2.778879165649414,
|
| 1487 |
+
"logits/rejected": -2.591799736022949,
|
| 1488 |
+
"logps/chosen": -103.28349304199219,
|
| 1489 |
+
"logps/rejected": -119.67398834228516,
|
| 1490 |
+
"loss": 0.4954,
|
| 1491 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1492 |
+
"rewards/chosen": -2.8379952907562256,
|
| 1493 |
+
"rewards/margins": 1.7630186080932617,
|
| 1494 |
+
"rewards/rejected": -4.601014137268066,
|
| 1495 |
+
"step": 990
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.7410151908114117,
|
| 1499 |
+
"grad_norm": 104.96907043457031,
|
| 1500 |
+
"learning_rate": 1.3222222222222222e-06,
|
| 1501 |
+
"logits/chosen": -2.5238616466522217,
|
| 1502 |
+
"logits/rejected": -2.4784162044525146,
|
| 1503 |
+
"logps/chosen": -107.37699890136719,
|
| 1504 |
+
"logps/rejected": -130.21548461914062,
|
| 1505 |
+
"loss": 0.5063,
|
| 1506 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1507 |
+
"rewards/chosen": -3.024723529815674,
|
| 1508 |
+
"rewards/margins": 1.954242467880249,
|
| 1509 |
+
"rewards/rejected": -4.978966236114502,
|
| 1510 |
+
"step": 1000
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 0.7484253427195258,
|
| 1514 |
+
"grad_norm": 44.657108306884766,
|
| 1515 |
+
"learning_rate": 1.2851851851851855e-06,
|
| 1516 |
+
"logits/chosen": -2.546823024749756,
|
| 1517 |
+
"logits/rejected": -2.4334805011749268,
|
| 1518 |
+
"logps/chosen": -106.6165771484375,
|
| 1519 |
+
"logps/rejected": -125.7170639038086,
|
| 1520 |
+
"loss": 0.3697,
|
| 1521 |
+
"rewards/accuracies": 0.8500000238418579,
|
| 1522 |
+
"rewards/chosen": -2.4411277770996094,
|
| 1523 |
+
"rewards/margins": 2.227128505706787,
|
| 1524 |
+
"rewards/rejected": -4.668255805969238,
|
| 1525 |
+
"step": 1010
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 0.7558354946276399,
|
| 1529 |
+
"grad_norm": 90.46170806884766,
|
| 1530 |
+
"learning_rate": 1.248148148148148e-06,
|
| 1531 |
+
"logits/chosen": -2.828085422515869,
|
| 1532 |
+
"logits/rejected": -2.6428349018096924,
|
| 1533 |
+
"logps/chosen": -105.6900405883789,
|
| 1534 |
+
"logps/rejected": -133.78445434570312,
|
| 1535 |
+
"loss": 0.3561,
|
| 1536 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1537 |
+
"rewards/chosen": -2.77301025390625,
|
| 1538 |
+
"rewards/margins": 2.5772953033447266,
|
| 1539 |
+
"rewards/rejected": -5.350305080413818,
|
| 1540 |
+
"step": 1020
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 0.763245646535754,
|
| 1544 |
+
"grad_norm": 21.936511993408203,
|
| 1545 |
+
"learning_rate": 1.2111111111111111e-06,
|
| 1546 |
+
"logits/chosen": -2.6992528438568115,
|
| 1547 |
+
"logits/rejected": -2.567366361618042,
|
| 1548 |
+
"logps/chosen": -103.77992248535156,
|
| 1549 |
+
"logps/rejected": -136.04214477539062,
|
| 1550 |
+
"loss": 0.405,
|
| 1551 |
+
"rewards/accuracies": 0.8500000238418579,
|
| 1552 |
+
"rewards/chosen": -3.2632954120635986,
|
| 1553 |
+
"rewards/margins": 2.4483699798583984,
|
| 1554 |
+
"rewards/rejected": -5.711665630340576,
|
| 1555 |
+
"step": 1030
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 0.7706557984438681,
|
| 1559 |
+
"grad_norm": 32.31972122192383,
|
| 1560 |
+
"learning_rate": 1.1740740740740742e-06,
|
| 1561 |
+
"logits/chosen": -2.772641897201538,
|
| 1562 |
+
"logits/rejected": -2.6311516761779785,
|
| 1563 |
+
"logps/chosen": -104.34123229980469,
|
| 1564 |
+
"logps/rejected": -128.68487548828125,
|
| 1565 |
+
"loss": 0.5004,
|
| 1566 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1567 |
+
"rewards/chosen": -3.079659938812256,
|
| 1568 |
+
"rewards/margins": 1.8224769830703735,
|
| 1569 |
+
"rewards/rejected": -4.90213680267334,
|
| 1570 |
+
"step": 1040
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7780659503519822,
|
| 1574 |
+
"grad_norm": 162.9198455810547,
|
| 1575 |
+
"learning_rate": 1.1370370370370372e-06,
|
| 1576 |
+
"logits/chosen": -2.8467555046081543,
|
| 1577 |
+
"logits/rejected": -2.515394926071167,
|
| 1578 |
+
"logps/chosen": -105.55623626708984,
|
| 1579 |
+
"logps/rejected": -125.57078552246094,
|
| 1580 |
+
"loss": 0.5177,
|
| 1581 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1582 |
+
"rewards/chosen": -3.2788283824920654,
|
| 1583 |
+
"rewards/margins": 1.9488775730133057,
|
| 1584 |
+
"rewards/rejected": -5.227705955505371,
|
| 1585 |
+
"step": 1050
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.7854761022600963,
|
| 1589 |
+
"grad_norm": 72.62833404541016,
|
| 1590 |
+
"learning_rate": 1.1e-06,
|
| 1591 |
+
"logits/chosen": -2.7466042041778564,
|
| 1592 |
+
"logits/rejected": -2.6573309898376465,
|
| 1593 |
+
"logps/chosen": -108.9305419921875,
|
| 1594 |
+
"logps/rejected": -129.41574096679688,
|
| 1595 |
+
"loss": 0.5488,
|
| 1596 |
+
"rewards/accuracies": 0.75,
|
| 1597 |
+
"rewards/chosen": -3.1458160877227783,
|
| 1598 |
+
"rewards/margins": 1.7604339122772217,
|
| 1599 |
+
"rewards/rejected": -4.906250476837158,
|
| 1600 |
+
"step": 1060
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.7928862541682105,
|
| 1604 |
+
"grad_norm": 33.611995697021484,
|
| 1605 |
+
"learning_rate": 1.062962962962963e-06,
|
| 1606 |
+
"logits/chosen": -2.7340753078460693,
|
| 1607 |
+
"logits/rejected": -2.536836862564087,
|
| 1608 |
+
"logps/chosen": -108.49119567871094,
|
| 1609 |
+
"logps/rejected": -133.3302764892578,
|
| 1610 |
+
"loss": 0.4382,
|
| 1611 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1612 |
+
"rewards/chosen": -3.197749614715576,
|
| 1613 |
+
"rewards/margins": 2.093479633331299,
|
| 1614 |
+
"rewards/rejected": -5.291229724884033,
|
| 1615 |
+
"step": 1070
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 0.8002964060763246,
|
| 1619 |
+
"grad_norm": 74.17855834960938,
|
| 1620 |
+
"learning_rate": 1.0259259259259261e-06,
|
| 1621 |
+
"logits/chosen": -2.8403737545013428,
|
| 1622 |
+
"logits/rejected": -2.6320414543151855,
|
| 1623 |
+
"logps/chosen": -114.62956237792969,
|
| 1624 |
+
"logps/rejected": -133.36578369140625,
|
| 1625 |
+
"loss": 0.4072,
|
| 1626 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 1627 |
+
"rewards/chosen": -3.09629487991333,
|
| 1628 |
+
"rewards/margins": 2.1655564308166504,
|
| 1629 |
+
"rewards/rejected": -5.2618513107299805,
|
| 1630 |
+
"step": 1080
|
| 1631 |
+
},
|
| 1632 |
+
{
|
| 1633 |
+
"epoch": 0.8077065579844387,
|
| 1634 |
+
"grad_norm": 54.3246955871582,
|
| 1635 |
+
"learning_rate": 9.88888888888889e-07,
|
| 1636 |
+
"logits/chosen": -2.557582139968872,
|
| 1637 |
+
"logits/rejected": -2.426109552383423,
|
| 1638 |
+
"logps/chosen": -110.915283203125,
|
| 1639 |
+
"logps/rejected": -134.53126525878906,
|
| 1640 |
+
"loss": 0.4985,
|
| 1641 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 1642 |
+
"rewards/chosen": -3.3661487102508545,
|
| 1643 |
+
"rewards/margins": 1.8312829732894897,
|
| 1644 |
+
"rewards/rejected": -5.197432041168213,
|
| 1645 |
+
"step": 1090
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 0.8151167098925528,
|
| 1649 |
+
"grad_norm": 33.4342041015625,
|
| 1650 |
+
"learning_rate": 9.51851851851852e-07,
|
| 1651 |
+
"logits/chosen": -2.843123197555542,
|
| 1652 |
+
"logits/rejected": -2.759000301361084,
|
| 1653 |
+
"logps/chosen": -112.1214828491211,
|
| 1654 |
+
"logps/rejected": -135.30357360839844,
|
| 1655 |
+
"loss": 0.4424,
|
| 1656 |
+
"rewards/accuracies": 0.7749999761581421,
|
| 1657 |
+
"rewards/chosen": -3.535928249359131,
|
| 1658 |
+
"rewards/margins": 1.8440746068954468,
|
| 1659 |
+
"rewards/rejected": -5.380003452301025,
|
| 1660 |
+
"step": 1100
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 0.8225268618006669,
|
| 1664 |
+
"grad_norm": 61.59471130371094,
|
| 1665 |
+
"learning_rate": 9.14814814814815e-07,
|
| 1666 |
+
"logits/chosen": -2.5715582370758057,
|
| 1667 |
+
"logits/rejected": -2.4673850536346436,
|
| 1668 |
+
"logps/chosen": -109.1755142211914,
|
| 1669 |
+
"logps/rejected": -132.7253875732422,
|
| 1670 |
+
"loss": 0.5435,
|
| 1671 |
+
"rewards/accuracies": 0.737500011920929,
|
| 1672 |
+
"rewards/chosen": -3.2600913047790527,
|
| 1673 |
+
"rewards/margins": 2.2205255031585693,
|
| 1674 |
+
"rewards/rejected": -5.480616569519043,
|
| 1675 |
+
"step": 1110
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.829937013708781,
|
| 1679 |
+
"grad_norm": 81.85456848144531,
|
| 1680 |
+
"learning_rate": 8.777777777777778e-07,
|
| 1681 |
+
"logits/chosen": -2.6088593006134033,
|
| 1682 |
+
"logits/rejected": -2.4452264308929443,
|
| 1683 |
+
"logps/chosen": -109.85011291503906,
|
| 1684 |
+
"logps/rejected": -135.8500518798828,
|
| 1685 |
+
"loss": 0.3276,
|
| 1686 |
+
"rewards/accuracies": 0.887499988079071,
|
| 1687 |
+
"rewards/chosen": -3.3149032592773438,
|
| 1688 |
+
"rewards/margins": 2.037301778793335,
|
| 1689 |
+
"rewards/rejected": -5.3522047996521,
|
| 1690 |
+
"step": 1120
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.8373471656168952,
|
| 1694 |
+
"grad_norm": 47.77298355102539,
|
| 1695 |
+
"learning_rate": 8.407407407407408e-07,
|
| 1696 |
+
"logits/chosen": -2.701817035675049,
|
| 1697 |
+
"logits/rejected": -2.6270086765289307,
|
| 1698 |
+
"logps/chosen": -108.1476058959961,
|
| 1699 |
+
"logps/rejected": -127.43824768066406,
|
| 1700 |
+
"loss": 0.5859,
|
| 1701 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 1702 |
+
"rewards/chosen": -3.2532172203063965,
|
| 1703 |
+
"rewards/margins": 1.7242540121078491,
|
| 1704 |
+
"rewards/rejected": -4.977471351623535,
|
| 1705 |
+
"step": 1130
|
| 1706 |
+
},
|
| 1707 |
+
{
|
| 1708 |
+
"epoch": 0.8447573175250093,
|
| 1709 |
+
"grad_norm": 70.82620239257812,
|
| 1710 |
+
"learning_rate": 8.037037037037038e-07,
|
| 1711 |
+
"logits/chosen": -2.738939046859741,
|
| 1712 |
+
"logits/rejected": -2.635751247406006,
|
| 1713 |
+
"logps/chosen": -101.56913757324219,
|
| 1714 |
+
"logps/rejected": -134.12655639648438,
|
| 1715 |
+
"loss": 0.3653,
|
| 1716 |
+
"rewards/accuracies": 0.824999988079071,
|
| 1717 |
+
"rewards/chosen": -2.848381280899048,
|
| 1718 |
+
"rewards/margins": 2.132460832595825,
|
| 1719 |
+
"rewards/rejected": -4.980842590332031,
|
| 1720 |
+
"step": 1140
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 0.8521674694331234,
|
| 1724 |
+
"grad_norm": 44.759246826171875,
|
| 1725 |
+
"learning_rate": 7.666666666666667e-07,
|
| 1726 |
+
"logits/chosen": -2.4769434928894043,
|
| 1727 |
+
"logits/rejected": -2.237217426300049,
|
| 1728 |
+
"logps/chosen": -103.843994140625,
|
| 1729 |
+
"logps/rejected": -129.95098876953125,
|
| 1730 |
+
"loss": 0.3671,
|
| 1731 |
+
"rewards/accuracies": 0.8500000238418579,
|
| 1732 |
+
"rewards/chosen": -3.2238991260528564,
|
| 1733 |
+
"rewards/margins": 2.1413819789886475,
|
| 1734 |
+
"rewards/rejected": -5.365281105041504,
|
| 1735 |
+
"step": 1150
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 0.8595776213412375,
|
| 1739 |
+
"grad_norm": 82.56806182861328,
|
| 1740 |
+
"learning_rate": 7.296296296296297e-07,
|
| 1741 |
+
"logits/chosen": -2.6925208568573,
|
| 1742 |
+
"logits/rejected": -2.4951748847961426,
|
| 1743 |
+
"logps/chosen": -108.81231689453125,
|
| 1744 |
+
"logps/rejected": -123.46539306640625,
|
| 1745 |
+
"loss": 0.5316,
|
| 1746 |
+
"rewards/accuracies": 0.75,
|
| 1747 |
+
"rewards/chosen": -2.8224823474884033,
|
| 1748 |
+
"rewards/margins": 1.9338245391845703,
|
| 1749 |
+
"rewards/rejected": -4.7563066482543945,
|
| 1750 |
+
"step": 1160
|
| 1751 |
+
},
|
| 1752 |
+
{
|
| 1753 |
+
"epoch": 0.8669877732493516,
|
| 1754 |
+
"grad_norm": 79.25946044921875,
|
| 1755 |
+
"learning_rate": 6.925925925925926e-07,
|
| 1756 |
+
"logits/chosen": -2.795814275741577,
|
| 1757 |
+
"logits/rejected": -2.6198534965515137,
|
| 1758 |
+
"logps/chosen": -114.69990539550781,
|
| 1759 |
+
"logps/rejected": -138.99383544921875,
|
| 1760 |
+
"loss": 0.4481,
|
| 1761 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1762 |
+
"rewards/chosen": -3.544638156890869,
|
| 1763 |
+
"rewards/margins": 1.8667469024658203,
|
| 1764 |
+
"rewards/rejected": -5.411385536193848,
|
| 1765 |
+
"step": 1170
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 0.8743979251574657,
|
| 1769 |
+
"grad_norm": 68.91136932373047,
|
| 1770 |
+
"learning_rate": 6.555555555555556e-07,
|
| 1771 |
+
"logits/chosen": -2.5401740074157715,
|
| 1772 |
+
"logits/rejected": -2.3985536098480225,
|
| 1773 |
+
"logps/chosen": -108.79698181152344,
|
| 1774 |
+
"logps/rejected": -127.04399108886719,
|
| 1775 |
+
"loss": 0.4939,
|
| 1776 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1777 |
+
"rewards/chosen": -3.1688778400421143,
|
| 1778 |
+
"rewards/margins": 1.6756236553192139,
|
| 1779 |
+
"rewards/rejected": -4.844501495361328,
|
| 1780 |
+
"step": 1180
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8818080770655798,
|
| 1784 |
+
"grad_norm": 153.0009765625,
|
| 1785 |
+
"learning_rate": 6.185185185185186e-07,
|
| 1786 |
+
"logits/chosen": -2.725766181945801,
|
| 1787 |
+
"logits/rejected": -2.6588385105133057,
|
| 1788 |
+
"logps/chosen": -115.24592590332031,
|
| 1789 |
+
"logps/rejected": -126.50578308105469,
|
| 1790 |
+
"loss": 0.6857,
|
| 1791 |
+
"rewards/accuracies": 0.637499988079071,
|
| 1792 |
+
"rewards/chosen": -3.7392802238464355,
|
| 1793 |
+
"rewards/margins": 1.0371185541152954,
|
| 1794 |
+
"rewards/rejected": -4.7763991355896,
|
| 1795 |
+
"step": 1190
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.889218228973694,
|
| 1799 |
+
"grad_norm": 50.05393981933594,
|
| 1800 |
+
"learning_rate": 5.814814814814816e-07,
|
| 1801 |
+
"logits/chosen": -2.800872325897217,
|
| 1802 |
+
"logits/rejected": -2.5388267040252686,
|
| 1803 |
+
"logps/chosen": -108.91355895996094,
|
| 1804 |
+
"logps/rejected": -140.00509643554688,
|
| 1805 |
+
"loss": 0.5076,
|
| 1806 |
+
"rewards/accuracies": 0.737500011920929,
|
| 1807 |
+
"rewards/chosen": -3.350496292114258,
|
| 1808 |
+
"rewards/margins": 2.106170892715454,
|
| 1809 |
+
"rewards/rejected": -5.456666946411133,
|
| 1810 |
+
"step": 1200
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.8966283808818081,
|
| 1814 |
+
"grad_norm": 67.96463775634766,
|
| 1815 |
+
"learning_rate": 5.444444444444444e-07,
|
| 1816 |
+
"logits/chosen": -2.7067558765411377,
|
| 1817 |
+
"logits/rejected": -2.599600315093994,
|
| 1818 |
+
"logps/chosen": -107.89814758300781,
|
| 1819 |
+
"logps/rejected": -124.91080474853516,
|
| 1820 |
+
"loss": 0.5518,
|
| 1821 |
+
"rewards/accuracies": 0.7124999761581421,
|
| 1822 |
+
"rewards/chosen": -3.1988513469696045,
|
| 1823 |
+
"rewards/margins": 1.6774219274520874,
|
| 1824 |
+
"rewards/rejected": -4.876273155212402,
|
| 1825 |
+
"step": 1210
|
| 1826 |
+
},
|
| 1827 |
+
{
|
| 1828 |
+
"epoch": 0.9040385327899222,
|
| 1829 |
+
"grad_norm": 51.324058532714844,
|
| 1830 |
+
"learning_rate": 5.074074074074075e-07,
|
| 1831 |
+
"logits/chosen": -2.5362696647644043,
|
| 1832 |
+
"logits/rejected": -2.354970693588257,
|
| 1833 |
+
"logps/chosen": -115.45835876464844,
|
| 1834 |
+
"logps/rejected": -136.94302368164062,
|
| 1835 |
+
"loss": 0.4526,
|
| 1836 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1837 |
+
"rewards/chosen": -3.65421986579895,
|
| 1838 |
+
"rewards/margins": 2.047398090362549,
|
| 1839 |
+
"rewards/rejected": -5.701618671417236,
|
| 1840 |
+
"step": 1220
|
| 1841 |
+
},
|
| 1842 |
+
{
|
| 1843 |
+
"epoch": 0.9114486846980363,
|
| 1844 |
+
"grad_norm": 72.69883728027344,
|
| 1845 |
+
"learning_rate": 4.703703703703704e-07,
|
| 1846 |
+
"logits/chosen": -2.776798725128174,
|
| 1847 |
+
"logits/rejected": -2.5926573276519775,
|
| 1848 |
+
"logps/chosen": -109.353515625,
|
| 1849 |
+
"logps/rejected": -133.78953552246094,
|
| 1850 |
+
"loss": 0.479,
|
| 1851 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1852 |
+
"rewards/chosen": -3.6324219703674316,
|
| 1853 |
+
"rewards/margins": 1.8643248081207275,
|
| 1854 |
+
"rewards/rejected": -5.496747016906738,
|
| 1855 |
+
"step": 1230
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 0.9188588366061504,
|
| 1859 |
+
"grad_norm": 76.83854675292969,
|
| 1860 |
+
"learning_rate": 4.333333333333334e-07,
|
| 1861 |
+
"logits/chosen": -2.752346992492676,
|
| 1862 |
+
"logits/rejected": -2.629638195037842,
|
| 1863 |
+
"logps/chosen": -105.3023910522461,
|
| 1864 |
+
"logps/rejected": -126.5351791381836,
|
| 1865 |
+
"loss": 0.5053,
|
| 1866 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1867 |
+
"rewards/chosen": -3.197103977203369,
|
| 1868 |
+
"rewards/margins": 1.733891487121582,
|
| 1869 |
+
"rewards/rejected": -4.930995464324951,
|
| 1870 |
+
"step": 1240
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 0.9262689885142645,
|
| 1874 |
+
"grad_norm": 110.72493743896484,
|
| 1875 |
+
"learning_rate": 3.9629629629629634e-07,
|
| 1876 |
+
"logits/chosen": -2.612422227859497,
|
| 1877 |
+
"logits/rejected": -2.4569220542907715,
|
| 1878 |
+
"logps/chosen": -97.88721466064453,
|
| 1879 |
+
"logps/rejected": -128.8172607421875,
|
| 1880 |
+
"loss": 0.4024,
|
| 1881 |
+
"rewards/accuracies": 0.800000011920929,
|
| 1882 |
+
"rewards/chosen": -3.1734538078308105,
|
| 1883 |
+
"rewards/margins": 2.2661962509155273,
|
| 1884 |
+
"rewards/rejected": -5.439650535583496,
|
| 1885 |
+
"step": 1250
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9336791404223787,
|
| 1889 |
+
"grad_norm": 79.09251403808594,
|
| 1890 |
+
"learning_rate": 3.592592592592593e-07,
|
| 1891 |
+
"logits/chosen": -2.5835788249969482,
|
| 1892 |
+
"logits/rejected": -2.425140857696533,
|
| 1893 |
+
"logps/chosen": -118.10221862792969,
|
| 1894 |
+
"logps/rejected": -140.7429656982422,
|
| 1895 |
+
"loss": 0.374,
|
| 1896 |
+
"rewards/accuracies": 0.8125,
|
| 1897 |
+
"rewards/chosen": -3.600130558013916,
|
| 1898 |
+
"rewards/margins": 2.1908457279205322,
|
| 1899 |
+
"rewards/rejected": -5.790975570678711,
|
| 1900 |
+
"step": 1260
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.9410892923304928,
|
| 1904 |
+
"grad_norm": 77.63102722167969,
|
| 1905 |
+
"learning_rate": 3.2222222222222227e-07,
|
| 1906 |
+
"logits/chosen": -2.833207607269287,
|
| 1907 |
+
"logits/rejected": -2.741405487060547,
|
| 1908 |
+
"logps/chosen": -105.04319763183594,
|
| 1909 |
+
"logps/rejected": -129.08157348632812,
|
| 1910 |
+
"loss": 0.4978,
|
| 1911 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1912 |
+
"rewards/chosen": -3.2051002979278564,
|
| 1913 |
+
"rewards/margins": 2.0284972190856934,
|
| 1914 |
+
"rewards/rejected": -5.233597278594971,
|
| 1915 |
+
"step": 1270
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 0.9484994442386069,
|
| 1919 |
+
"grad_norm": 45.666744232177734,
|
| 1920 |
+
"learning_rate": 2.851851851851852e-07,
|
| 1921 |
+
"logits/chosen": -2.8316311836242676,
|
| 1922 |
+
"logits/rejected": -2.6417407989501953,
|
| 1923 |
+
"logps/chosen": -113.1464614868164,
|
| 1924 |
+
"logps/rejected": -136.09994506835938,
|
| 1925 |
+
"loss": 0.3906,
|
| 1926 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1927 |
+
"rewards/chosen": -3.425858736038208,
|
| 1928 |
+
"rewards/margins": 2.1212241649627686,
|
| 1929 |
+
"rewards/rejected": -5.547082424163818,
|
| 1930 |
+
"step": 1280
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 0.955909596146721,
|
| 1934 |
+
"grad_norm": 126.80838775634766,
|
| 1935 |
+
"learning_rate": 2.4814814814814814e-07,
|
| 1936 |
+
"logits/chosen": -2.479860782623291,
|
| 1937 |
+
"logits/rejected": -2.4349493980407715,
|
| 1938 |
+
"logps/chosen": -110.22750091552734,
|
| 1939 |
+
"logps/rejected": -130.07176208496094,
|
| 1940 |
+
"loss": 0.5546,
|
| 1941 |
+
"rewards/accuracies": 0.6875,
|
| 1942 |
+
"rewards/chosen": -3.495988368988037,
|
| 1943 |
+
"rewards/margins": 1.9390910863876343,
|
| 1944 |
+
"rewards/rejected": -5.435080051422119,
|
| 1945 |
+
"step": 1290
|
| 1946 |
+
},
|
| 1947 |
+
{
|
| 1948 |
+
"epoch": 0.9633197480548351,
|
| 1949 |
+
"grad_norm": 104.56671142578125,
|
| 1950 |
+
"learning_rate": 2.1111111111111113e-07,
|
| 1951 |
+
"logits/chosen": -2.8810842037200928,
|
| 1952 |
+
"logits/rejected": -2.64825701713562,
|
| 1953 |
+
"logps/chosen": -114.70051574707031,
|
| 1954 |
+
"logps/rejected": -133.77554321289062,
|
| 1955 |
+
"loss": 0.4823,
|
| 1956 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1957 |
+
"rewards/chosen": -3.2450995445251465,
|
| 1958 |
+
"rewards/margins": 1.894256830215454,
|
| 1959 |
+
"rewards/rejected": -5.13935661315918,
|
| 1960 |
+
"step": 1300
|
| 1961 |
+
},
|
| 1962 |
+
{
|
| 1963 |
+
"epoch": 0.9707298999629492,
|
| 1964 |
+
"grad_norm": 69.64128875732422,
|
| 1965 |
+
"learning_rate": 1.7407407407407407e-07,
|
| 1966 |
+
"logits/chosen": -2.7370734214782715,
|
| 1967 |
+
"logits/rejected": -2.5394835472106934,
|
| 1968 |
+
"logps/chosen": -120.4969711303711,
|
| 1969 |
+
"logps/rejected": -136.19540405273438,
|
| 1970 |
+
"loss": 0.6125,
|
| 1971 |
+
"rewards/accuracies": 0.762499988079071,
|
| 1972 |
+
"rewards/chosen": -3.9360873699188232,
|
| 1973 |
+
"rewards/margins": 1.48752760887146,
|
| 1974 |
+
"rewards/rejected": -5.423615455627441,
|
| 1975 |
+
"step": 1310
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 0.9781400518710633,
|
| 1979 |
+
"grad_norm": 72.55949401855469,
|
| 1980 |
+
"learning_rate": 1.3703703703703706e-07,
|
| 1981 |
+
"logits/chosen": -2.5739331245422363,
|
| 1982 |
+
"logits/rejected": -2.3697855472564697,
|
| 1983 |
+
"logps/chosen": -111.98551940917969,
|
| 1984 |
+
"logps/rejected": -126.80464935302734,
|
| 1985 |
+
"loss": 0.4807,
|
| 1986 |
+
"rewards/accuracies": 0.7250000238418579,
|
| 1987 |
+
"rewards/chosen": -3.24163556098938,
|
| 1988 |
+
"rewards/margins": 1.6213098764419556,
|
| 1989 |
+
"rewards/rejected": -4.862946510314941,
|
| 1990 |
+
"step": 1320
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9855502037791775,
|
| 1994 |
+
"grad_norm": 74.78096008300781,
|
| 1995 |
+
"learning_rate": 1.0000000000000001e-07,
|
| 1996 |
+
"logits/chosen": -2.6215052604675293,
|
| 1997 |
+
"logits/rejected": -2.4891343116760254,
|
| 1998 |
+
"logps/chosen": -104.3646469116211,
|
| 1999 |
+
"logps/rejected": -129.4760284423828,
|
| 2000 |
+
"loss": 0.377,
|
| 2001 |
+
"rewards/accuracies": 0.8999999761581421,
|
| 2002 |
+
"rewards/chosen": -2.8517489433288574,
|
| 2003 |
+
"rewards/margins": 2.320369005203247,
|
| 2004 |
+
"rewards/rejected": -5.172118186950684,
|
| 2005 |
+
"step": 1330
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.9929603556872916,
|
| 2009 |
+
"grad_norm": 88.45356750488281,
|
| 2010 |
+
"learning_rate": 6.296296296296297e-08,
|
| 2011 |
+
"logits/chosen": -2.569523334503174,
|
| 2012 |
+
"logits/rejected": -2.4072022438049316,
|
| 2013 |
+
"logps/chosen": -110.60871887207031,
|
| 2014 |
+
"logps/rejected": -127.8760986328125,
|
| 2015 |
+
"loss": 0.5243,
|
| 2016 |
+
"rewards/accuracies": 0.737500011920929,
|
| 2017 |
+
"rewards/chosen": -3.452281951904297,
|
| 2018 |
+
"rewards/margins": 1.5607750415802002,
|
| 2019 |
+
"rewards/rejected": -5.013056755065918,
|
| 2020 |
+
"step": 1340
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 1.0,
|
| 2024 |
+
"grad_norm": 10.505631446838379,
|
| 2025 |
+
"learning_rate": 2.5925925925925926e-08,
|
| 2026 |
+
"logits/chosen": -2.393158435821533,
|
| 2027 |
+
"logits/rejected": -2.35024356842041,
|
| 2028 |
+
"logps/chosen": -112.46697998046875,
|
| 2029 |
+
"logps/rejected": -128.6726837158203,
|
| 2030 |
+
"loss": 0.4409,
|
| 2031 |
+
"rewards/accuracies": 0.7894737124443054,
|
| 2032 |
+
"rewards/chosen": -3.711907148361206,
|
| 2033 |
+
"rewards/margins": 1.5488325357437134,
|
| 2034 |
+
"rewards/rejected": -5.260739803314209,
|
| 2035 |
+
"step": 1350
|
| 2036 |
+
}
|
| 2037 |
+
],
|
| 2038 |
+
"logging_steps": 10,
|
| 2039 |
+
"max_steps": 1350,
|
| 2040 |
+
"num_input_tokens_seen": 0,
|
| 2041 |
+
"num_train_epochs": 1,
|
| 2042 |
+
"save_steps": 500,
|
| 2043 |
+
"stateful_callbacks": {
|
| 2044 |
+
"TrainerControl": {
|
| 2045 |
+
"args": {
|
| 2046 |
+
"should_epoch_stop": false,
|
| 2047 |
+
"should_evaluate": false,
|
| 2048 |
+
"should_log": false,
|
| 2049 |
+
"should_save": true,
|
| 2050 |
+
"should_training_stop": true
|
| 2051 |
+
},
|
| 2052 |
+
"attributes": {}
|
| 2053 |
+
}
|
| 2054 |
+
},
|
| 2055 |
+
"total_flos": 0.0,
|
| 2056 |
+
"train_batch_size": 2,
|
| 2057 |
+
"trial_name": null,
|
| 2058 |
+
"trial_params": null
|
| 2059 |
+
}
|
checkpoint-1350/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3a7b7968a8ae2ad676d8cb9afe1a793e119b7999c5723f8eae7378c1356c231
|
| 3 |
+
size 6392
|
checkpoint-1350/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"Qwen3ForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 151643,
|
| 8 |
+
"eos_token_id": 151643,
|
| 9 |
+
"head_dim": 128,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 1024,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"max_position_embeddings": 32768,
|
| 15 |
+
"max_window_layers": 28,
|
| 16 |
+
"model_type": "qwen3",
|
| 17 |
+
"num_attention_heads": 16,
|
| 18 |
+
"num_hidden_layers": 28,
|
| 19 |
+
"num_key_value_heads": 8,
|
| 20 |
+
"rms_norm_eps": 1e-06,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rope_theta": 1000000,
|
| 23 |
+
"sliding_window": null,
|
| 24 |
+
"tie_word_embeddings": true,
|
| 25 |
+
"torch_dtype": "float32",
|
| 26 |
+
"transformers_version": "4.52.3",
|
| 27 |
+
"use_cache": true,
|
| 28 |
+
"use_sliding_window": false,
|
| 29 |
+
"vocab_size": 151936
|
| 30 |
+
}
|
dpo_config.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"beta": 0.1,
|
| 3 |
+
"learning_rate": 5e-06,
|
| 4 |
+
"per_device_train_batch_size": 2,
|
| 5 |
+
"gradient_accumulation_steps": 4,
|
| 6 |
+
"max_length": 512,
|
| 7 |
+
"max_prompt_length": 128,
|
| 8 |
+
"num_train_epochs": 1,
|
| 9 |
+
"logging_steps": 10,
|
| 10 |
+
"save_strategy": "epoch",
|
| 11 |
+
"output_dir": "./koreankiwi99_dpo_model_base_Math-Step-DPO-10K",
|
| 12 |
+
"remove_unused_columns": false,
|
| 13 |
+
"fp16": true,
|
| 14 |
+
"bf16": false,
|
| 15 |
+
"gradient_checkpointing": false,
|
| 16 |
+
"max_grad_norm": 1.0,
|
| 17 |
+
"push_to_hub_model_id": "koreankiwi99/dpo_model_base_Math-Step-DPO-10K"
|
| 18 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"eos_token_id": 151643,
|
| 4 |
+
"max_new_tokens": 2048,
|
| 5 |
+
"transformers_version": "4.52.3"
|
| 6 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a793c47676550a51ebca3f46d38c9caf6f35ccebbc241c469e660d591a568bc0
|
| 3 |
+
size 2384234968
|
runs/Jun05_14-17-48_47dafa9566a9/events.out.tfevents.1749133094.47dafa9566a9.1112.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d1decf67b86a4524ea49d67ed4a789a9486fb37d14d9d98f8e93acf00754f0a
|
| 3 |
+
size 99192
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|object_ref_start|>",
|
| 6 |
+
"<|object_ref_end|>",
|
| 7 |
+
"<|box_start|>",
|
| 8 |
+
"<|box_end|>",
|
| 9 |
+
"<|quad_start|>",
|
| 10 |
+
"<|quad_end|>",
|
| 11 |
+
"<|vision_start|>",
|
| 12 |
+
"<|vision_end|>",
|
| 13 |
+
"<|vision_pad|>",
|
| 14 |
+
"<|image_pad|>",
|
| 15 |
+
"<|video_pad|>"
|
| 16 |
+
],
|
| 17 |
+
"eos_token": {
|
| 18 |
+
"content": "<|endoftext|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
"pad_token": "<|endoftext|>"
|
| 25 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
|
| 3 |
+
size 11422654
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,240 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"151643": {
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"151644": {
|
| 14 |
+
"content": "<|im_start|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"151645": {
|
| 22 |
+
"content": "<|im_end|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"151646": {
|
| 30 |
+
"content": "<|object_ref_start|>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"151647": {
|
| 38 |
+
"content": "<|object_ref_end|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"151648": {
|
| 46 |
+
"content": "<|box_start|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
},
|
| 53 |
+
"151649": {
|
| 54 |
+
"content": "<|box_end|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false,
|
| 59 |
+
"special": true
|
| 60 |
+
},
|
| 61 |
+
"151650": {
|
| 62 |
+
"content": "<|quad_start|>",
|
| 63 |
+
"lstrip": false,
|
| 64 |
+
"normalized": false,
|
| 65 |
+
"rstrip": false,
|
| 66 |
+
"single_word": false,
|
| 67 |
+
"special": true
|
| 68 |
+
},
|
| 69 |
+
"151651": {
|
| 70 |
+
"content": "<|quad_end|>",
|
| 71 |
+
"lstrip": false,
|
| 72 |
+
"normalized": false,
|
| 73 |
+
"rstrip": false,
|
| 74 |
+
"single_word": false,
|
| 75 |
+
"special": true
|
| 76 |
+
},
|
| 77 |
+
"151652": {
|
| 78 |
+
"content": "<|vision_start|>",
|
| 79 |
+
"lstrip": false,
|
| 80 |
+
"normalized": false,
|
| 81 |
+
"rstrip": false,
|
| 82 |
+
"single_word": false,
|
| 83 |
+
"special": true
|
| 84 |
+
},
|
| 85 |
+
"151653": {
|
| 86 |
+
"content": "<|vision_end|>",
|
| 87 |
+
"lstrip": false,
|
| 88 |
+
"normalized": false,
|
| 89 |
+
"rstrip": false,
|
| 90 |
+
"single_word": false,
|
| 91 |
+
"special": true
|
| 92 |
+
},
|
| 93 |
+
"151654": {
|
| 94 |
+
"content": "<|vision_pad|>",
|
| 95 |
+
"lstrip": false,
|
| 96 |
+
"normalized": false,
|
| 97 |
+
"rstrip": false,
|
| 98 |
+
"single_word": false,
|
| 99 |
+
"special": true
|
| 100 |
+
},
|
| 101 |
+
"151655": {
|
| 102 |
+
"content": "<|image_pad|>",
|
| 103 |
+
"lstrip": false,
|
| 104 |
+
"normalized": false,
|
| 105 |
+
"rstrip": false,
|
| 106 |
+
"single_word": false,
|
| 107 |
+
"special": true
|
| 108 |
+
},
|
| 109 |
+
"151656": {
|
| 110 |
+
"content": "<|video_pad|>",
|
| 111 |
+
"lstrip": false,
|
| 112 |
+
"normalized": false,
|
| 113 |
+
"rstrip": false,
|
| 114 |
+
"single_word": false,
|
| 115 |
+
"special": true
|
| 116 |
+
},
|
| 117 |
+
"151657": {
|
| 118 |
+
"content": "<tool_call>",
|
| 119 |
+
"lstrip": false,
|
| 120 |
+
"normalized": false,
|
| 121 |
+
"rstrip": false,
|
| 122 |
+
"single_word": false,
|
| 123 |
+
"special": false
|
| 124 |
+
},
|
| 125 |
+
"151658": {
|
| 126 |
+
"content": "</tool_call>",
|
| 127 |
+
"lstrip": false,
|
| 128 |
+
"normalized": false,
|
| 129 |
+
"rstrip": false,
|
| 130 |
+
"single_word": false,
|
| 131 |
+
"special": false
|
| 132 |
+
},
|
| 133 |
+
"151659": {
|
| 134 |
+
"content": "<|fim_prefix|>",
|
| 135 |
+
"lstrip": false,
|
| 136 |
+
"normalized": false,
|
| 137 |
+
"rstrip": false,
|
| 138 |
+
"single_word": false,
|
| 139 |
+
"special": false
|
| 140 |
+
},
|
| 141 |
+
"151660": {
|
| 142 |
+
"content": "<|fim_middle|>",
|
| 143 |
+
"lstrip": false,
|
| 144 |
+
"normalized": false,
|
| 145 |
+
"rstrip": false,
|
| 146 |
+
"single_word": false,
|
| 147 |
+
"special": false
|
| 148 |
+
},
|
| 149 |
+
"151661": {
|
| 150 |
+
"content": "<|fim_suffix|>",
|
| 151 |
+
"lstrip": false,
|
| 152 |
+
"normalized": false,
|
| 153 |
+
"rstrip": false,
|
| 154 |
+
"single_word": false,
|
| 155 |
+
"special": false
|
| 156 |
+
},
|
| 157 |
+
"151662": {
|
| 158 |
+
"content": "<|fim_pad|>",
|
| 159 |
+
"lstrip": false,
|
| 160 |
+
"normalized": false,
|
| 161 |
+
"rstrip": false,
|
| 162 |
+
"single_word": false,
|
| 163 |
+
"special": false
|
| 164 |
+
},
|
| 165 |
+
"151663": {
|
| 166 |
+
"content": "<|repo_name|>",
|
| 167 |
+
"lstrip": false,
|
| 168 |
+
"normalized": false,
|
| 169 |
+
"rstrip": false,
|
| 170 |
+
"single_word": false,
|
| 171 |
+
"special": false
|
| 172 |
+
},
|
| 173 |
+
"151664": {
|
| 174 |
+
"content": "<|file_sep|>",
|
| 175 |
+
"lstrip": false,
|
| 176 |
+
"normalized": false,
|
| 177 |
+
"rstrip": false,
|
| 178 |
+
"single_word": false,
|
| 179 |
+
"special": false
|
| 180 |
+
},
|
| 181 |
+
"151665": {
|
| 182 |
+
"content": "<tool_response>",
|
| 183 |
+
"lstrip": false,
|
| 184 |
+
"normalized": false,
|
| 185 |
+
"rstrip": false,
|
| 186 |
+
"single_word": false,
|
| 187 |
+
"special": false
|
| 188 |
+
},
|
| 189 |
+
"151666": {
|
| 190 |
+
"content": "</tool_response>",
|
| 191 |
+
"lstrip": false,
|
| 192 |
+
"normalized": false,
|
| 193 |
+
"rstrip": false,
|
| 194 |
+
"single_word": false,
|
| 195 |
+
"special": false
|
| 196 |
+
},
|
| 197 |
+
"151667": {
|
| 198 |
+
"content": "<think>",
|
| 199 |
+
"lstrip": false,
|
| 200 |
+
"normalized": false,
|
| 201 |
+
"rstrip": false,
|
| 202 |
+
"single_word": false,
|
| 203 |
+
"special": false
|
| 204 |
+
},
|
| 205 |
+
"151668": {
|
| 206 |
+
"content": "</think>",
|
| 207 |
+
"lstrip": false,
|
| 208 |
+
"normalized": false,
|
| 209 |
+
"rstrip": false,
|
| 210 |
+
"single_word": false,
|
| 211 |
+
"special": false
|
| 212 |
+
}
|
| 213 |
+
},
|
| 214 |
+
"additional_special_tokens": [
|
| 215 |
+
"<|im_start|>",
|
| 216 |
+
"<|im_end|>",
|
| 217 |
+
"<|object_ref_start|>",
|
| 218 |
+
"<|object_ref_end|>",
|
| 219 |
+
"<|box_start|>",
|
| 220 |
+
"<|box_end|>",
|
| 221 |
+
"<|quad_start|>",
|
| 222 |
+
"<|quad_end|>",
|
| 223 |
+
"<|vision_start|>",
|
| 224 |
+
"<|vision_end|>",
|
| 225 |
+
"<|vision_pad|>",
|
| 226 |
+
"<|image_pad|>",
|
| 227 |
+
"<|video_pad|>"
|
| 228 |
+
],
|
| 229 |
+
"bos_token": null,
|
| 230 |
+
"chat_template": null,
|
| 231 |
+
"clean_up_tokenization_spaces": false,
|
| 232 |
+
"eos_token": "<|endoftext|>",
|
| 233 |
+
"errors": "replace",
|
| 234 |
+
"extra_special_tokens": {},
|
| 235 |
+
"model_max_length": 131072,
|
| 236 |
+
"pad_token": "<|endoftext|>",
|
| 237 |
+
"split_special_tokens": false,
|
| 238 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 239 |
+
"unk_token": null
|
| 240 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|