Upload Training Model
Browse files- test/checkpoint-200/config.json +39 -0
- test/checkpoint-200/generation_config.json +6 -0
- test/checkpoint-200/merges.txt +0 -0
- test/checkpoint-200/model.safetensors +3 -0
- test/checkpoint-200/optimizer.pt +3 -0
- test/checkpoint-200/rng_state.pth +3 -0
- test/checkpoint-200/scheduler.pt +3 -0
- test/checkpoint-200/special_tokens_map.json +6 -0
- test/checkpoint-200/tokenizer.json +0 -0
- test/checkpoint-200/tokenizer_config.json +20 -0
- test/checkpoint-200/trainer_state.json +648 -0
- test/checkpoint-200/training_args.bin +3 -0
- test/checkpoint-200/vocab.json +0 -0
- test/config.json +39 -0
- test/generation_config.json +6 -0
- test/merges.txt +0 -0
- test/model.safetensors +3 -0
- test/runs/Feb28_18-34-47_puffer/events.out.tfevents.1740767688.puffer.3556740.1 +3 -0
- test/special_tokens_map.json +6 -0
- test/tokenizer.json +0 -0
- test/tokenizer_config.json +20 -0
- test/training_args.bin +3 -0
- test/vocab.json +0 -0
test/checkpoint-200/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": false,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.46.0",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50257
|
| 39 |
+
}
|
test/checkpoint-200/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.46.0"
|
| 6 |
+
}
|
test/checkpoint-200/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
test/checkpoint-200/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b9cd4ea6d70556b7f7a032265263517f7730718cfafa3e7b911b4010c650282
|
| 3 |
+
size 497774208
|
test/checkpoint-200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a676ee5f6593b8f0ba2b94bece2144801385d9f2084d9d474c9827fa2b590ce
|
| 3 |
+
size 497804154
|
test/checkpoint-200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
|
| 3 |
+
size 14244
|
test/checkpoint-200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c51bc01e0edf1a4262e736512af0ae0311d9ae97b6846f516c790f2a5842b43c
|
| 3 |
+
size 1064
|
test/checkpoint-200/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
test/checkpoint-200/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
test/checkpoint-200/tokenizer_config.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
}
|
| 12 |
+
},
|
| 13 |
+
"bos_token": "<|endoftext|>",
|
| 14 |
+
"clean_up_tokenization_spaces": false,
|
| 15 |
+
"eos_token": "<|endoftext|>",
|
| 16 |
+
"model_max_length": 1024,
|
| 17 |
+
"pad_token": "<|endoftext|>",
|
| 18 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 19 |
+
"unk_token": "<|endoftext|>"
|
| 20 |
+
}
|
test/checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,648 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.6,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.008,
|
| 13 |
+
"grad_norm": 26.318634033203125,
|
| 14 |
+
"learning_rate": 6.6666666666666675e-06,
|
| 15 |
+
"logits/chosen": -122.30106353759766,
|
| 16 |
+
"logits/rejected": -119.91307067871094,
|
| 17 |
+
"logps/chosen": -44.68114471435547,
|
| 18 |
+
"logps/rejected": -63.82457733154297,
|
| 19 |
+
"loss": 0.6931,
|
| 20 |
+
"rewards/accuracies": 0.0,
|
| 21 |
+
"rewards/chosen": 0.0,
|
| 22 |
+
"rewards/margins": 0.0,
|
| 23 |
+
"rewards/rejected": 0.0,
|
| 24 |
+
"step": 1
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.04,
|
| 28 |
+
"grad_norm": 50.06523895263672,
|
| 29 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 30 |
+
"logits/chosen": -127.76481628417969,
|
| 31 |
+
"logits/rejected": -116.89424133300781,
|
| 32 |
+
"logps/chosen": -52.29368209838867,
|
| 33 |
+
"logps/rejected": -35.88545227050781,
|
| 34 |
+
"loss": 0.8326,
|
| 35 |
+
"rewards/accuracies": 0.21875,
|
| 36 |
+
"rewards/chosen": -0.3712196350097656,
|
| 37 |
+
"rewards/margins": -0.13846339285373688,
|
| 38 |
+
"rewards/rejected": -0.23275624215602875,
|
| 39 |
+
"step": 5
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.08,
|
| 43 |
+
"grad_norm": 11.627303123474121,
|
| 44 |
+
"learning_rate": 6.666666666666667e-05,
|
| 45 |
+
"logits/chosen": -108.85810852050781,
|
| 46 |
+
"logits/rejected": -109.94535064697266,
|
| 47 |
+
"logps/chosen": -63.54102325439453,
|
| 48 |
+
"logps/rejected": -85.88517761230469,
|
| 49 |
+
"loss": 0.8937,
|
| 50 |
+
"rewards/accuracies": 0.32499998807907104,
|
| 51 |
+
"rewards/chosen": -0.84485924243927,
|
| 52 |
+
"rewards/margins": -0.032330285757780075,
|
| 53 |
+
"rewards/rejected": -0.8125289082527161,
|
| 54 |
+
"step": 10
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.12,
|
| 58 |
+
"grad_norm": 57.29490280151367,
|
| 59 |
+
"learning_rate": 0.0001,
|
| 60 |
+
"logits/chosen": -109.92794036865234,
|
| 61 |
+
"logits/rejected": -106.7579574584961,
|
| 62 |
+
"logps/chosen": -81.18146514892578,
|
| 63 |
+
"logps/rejected": -68.21682739257812,
|
| 64 |
+
"loss": 1.3532,
|
| 65 |
+
"rewards/accuracies": 0.25,
|
| 66 |
+
"rewards/chosen": -1.8736642599105835,
|
| 67 |
+
"rewards/margins": -0.5895371437072754,
|
| 68 |
+
"rewards/rejected": -1.284126877784729,
|
| 69 |
+
"step": 15
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.16,
|
| 73 |
+
"grad_norm": 13.233222961425781,
|
| 74 |
+
"learning_rate": 0.00013333333333333334,
|
| 75 |
+
"logits/chosen": -102.33721923828125,
|
| 76 |
+
"logits/rejected": -98.88597106933594,
|
| 77 |
+
"logps/chosen": -61.5162353515625,
|
| 78 |
+
"logps/rejected": -75.63683319091797,
|
| 79 |
+
"loss": 0.7919,
|
| 80 |
+
"rewards/accuracies": 0.3499999940395355,
|
| 81 |
+
"rewards/chosen": -0.6732734441757202,
|
| 82 |
+
"rewards/margins": 0.4135725498199463,
|
| 83 |
+
"rewards/rejected": -1.0868459939956665,
|
| 84 |
+
"step": 20
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.2,
|
| 88 |
+
"grad_norm": 14.27536392211914,
|
| 89 |
+
"learning_rate": 0.00016666666666666666,
|
| 90 |
+
"logits/chosen": -81.1920394897461,
|
| 91 |
+
"logits/rejected": -81.63721466064453,
|
| 92 |
+
"logps/chosen": -63.35576248168945,
|
| 93 |
+
"logps/rejected": -78.37104797363281,
|
| 94 |
+
"loss": 0.7626,
|
| 95 |
+
"rewards/accuracies": 0.5,
|
| 96 |
+
"rewards/chosen": -1.5874093770980835,
|
| 97 |
+
"rewards/margins": 0.9996956586837769,
|
| 98 |
+
"rewards/rejected": -2.5871052742004395,
|
| 99 |
+
"step": 25
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.24,
|
| 103 |
+
"grad_norm": 20.265674591064453,
|
| 104 |
+
"learning_rate": 0.0002,
|
| 105 |
+
"logits/chosen": -63.422760009765625,
|
| 106 |
+
"logits/rejected": -59.9358024597168,
|
| 107 |
+
"logps/chosen": -77.57817077636719,
|
| 108 |
+
"logps/rejected": -110.3768081665039,
|
| 109 |
+
"loss": 0.749,
|
| 110 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 111 |
+
"rewards/chosen": -2.5955891609191895,
|
| 112 |
+
"rewards/margins": 1.2783219814300537,
|
| 113 |
+
"rewards/rejected": -3.8739113807678223,
|
| 114 |
+
"step": 30
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.28,
|
| 118 |
+
"grad_norm": 20.7922420501709,
|
| 119 |
+
"learning_rate": 0.00023333333333333333,
|
| 120 |
+
"logits/chosen": -62.76293182373047,
|
| 121 |
+
"logits/rejected": -60.520294189453125,
|
| 122 |
+
"logps/chosen": -125.72471618652344,
|
| 123 |
+
"logps/rejected": -121.91011810302734,
|
| 124 |
+
"loss": 2.0564,
|
| 125 |
+
"rewards/accuracies": 0.2750000059604645,
|
| 126 |
+
"rewards/chosen": -5.747698783874512,
|
| 127 |
+
"rewards/margins": -0.41697731614112854,
|
| 128 |
+
"rewards/rejected": -5.330721855163574,
|
| 129 |
+
"step": 35
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.32,
|
| 133 |
+
"grad_norm": 18.06304359436035,
|
| 134 |
+
"learning_rate": 0.0002666666666666667,
|
| 135 |
+
"logits/chosen": -53.0257453918457,
|
| 136 |
+
"logits/rejected": -53.412635803222656,
|
| 137 |
+
"logps/chosen": -82.18067932128906,
|
| 138 |
+
"logps/rejected": -94.44306182861328,
|
| 139 |
+
"loss": 1.4246,
|
| 140 |
+
"rewards/accuracies": 0.3499999940395355,
|
| 141 |
+
"rewards/chosen": -2.327519178390503,
|
| 142 |
+
"rewards/margins": 0.1818149834871292,
|
| 143 |
+
"rewards/rejected": -2.509334087371826,
|
| 144 |
+
"step": 40
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.36,
|
| 148 |
+
"grad_norm": 12.843459129333496,
|
| 149 |
+
"learning_rate": 0.0003,
|
| 150 |
+
"logits/chosen": -86.56561279296875,
|
| 151 |
+
"logits/rejected": -87.3717269897461,
|
| 152 |
+
"logps/chosen": -115.32487487792969,
|
| 153 |
+
"logps/rejected": -160.50839233398438,
|
| 154 |
+
"loss": 1.1277,
|
| 155 |
+
"rewards/accuracies": 0.375,
|
| 156 |
+
"rewards/chosen": -6.436800479888916,
|
| 157 |
+
"rewards/margins": 2.7378334999084473,
|
| 158 |
+
"rewards/rejected": -9.174633979797363,
|
| 159 |
+
"step": 45
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.4,
|
| 163 |
+
"grad_norm": 31.90726089477539,
|
| 164 |
+
"learning_rate": 0.0003333333333333333,
|
| 165 |
+
"logits/chosen": -63.01441192626953,
|
| 166 |
+
"logits/rejected": -63.63895797729492,
|
| 167 |
+
"logps/chosen": -122.78414154052734,
|
| 168 |
+
"logps/rejected": -120.47871398925781,
|
| 169 |
+
"loss": 2.112,
|
| 170 |
+
"rewards/accuracies": 0.3499999940395355,
|
| 171 |
+
"rewards/chosen": -6.3497443199157715,
|
| 172 |
+
"rewards/margins": 0.06578154861927032,
|
| 173 |
+
"rewards/rejected": -6.415525913238525,
|
| 174 |
+
"step": 50
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.44,
|
| 178 |
+
"grad_norm": 6.040111064910889,
|
| 179 |
+
"learning_rate": 0.00036666666666666667,
|
| 180 |
+
"logits/chosen": -43.232513427734375,
|
| 181 |
+
"logits/rejected": -41.53527069091797,
|
| 182 |
+
"logps/chosen": -86.36963653564453,
|
| 183 |
+
"logps/rejected": -99.5210952758789,
|
| 184 |
+
"loss": 1.2432,
|
| 185 |
+
"rewards/accuracies": 0.4749999940395355,
|
| 186 |
+
"rewards/chosen": -2.8871474266052246,
|
| 187 |
+
"rewards/margins": 0.6384552717208862,
|
| 188 |
+
"rewards/rejected": -3.5256030559539795,
|
| 189 |
+
"step": 55
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.48,
|
| 193 |
+
"grad_norm": 8.189231872558594,
|
| 194 |
+
"learning_rate": 0.0004,
|
| 195 |
+
"logits/chosen": -42.83661651611328,
|
| 196 |
+
"logits/rejected": -45.68109130859375,
|
| 197 |
+
"logps/chosen": -76.9494857788086,
|
| 198 |
+
"logps/rejected": -113.4403305053711,
|
| 199 |
+
"loss": 1.0003,
|
| 200 |
+
"rewards/accuracies": 0.4000000059604645,
|
| 201 |
+
"rewards/chosen": -3.5278103351593018,
|
| 202 |
+
"rewards/margins": 2.096407413482666,
|
| 203 |
+
"rewards/rejected": -5.624217510223389,
|
| 204 |
+
"step": 60
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.52,
|
| 208 |
+
"grad_norm": 10.270423889160156,
|
| 209 |
+
"learning_rate": 0.00043333333333333337,
|
| 210 |
+
"logits/chosen": -41.67694091796875,
|
| 211 |
+
"logits/rejected": -41.690773010253906,
|
| 212 |
+
"logps/chosen": -88.95043182373047,
|
| 213 |
+
"logps/rejected": -140.64022827148438,
|
| 214 |
+
"loss": 0.7315,
|
| 215 |
+
"rewards/accuracies": 0.44999998807907104,
|
| 216 |
+
"rewards/chosen": -3.7248928546905518,
|
| 217 |
+
"rewards/margins": 2.995668411254883,
|
| 218 |
+
"rewards/rejected": -6.7205610275268555,
|
| 219 |
+
"step": 65
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.56,
|
| 223 |
+
"grad_norm": 13.594496726989746,
|
| 224 |
+
"learning_rate": 0.00046666666666666666,
|
| 225 |
+
"logits/chosen": -30.712650299072266,
|
| 226 |
+
"logits/rejected": -29.960418701171875,
|
| 227 |
+
"logps/chosen": -91.05331420898438,
|
| 228 |
+
"logps/rejected": -123.3448486328125,
|
| 229 |
+
"loss": 1.1866,
|
| 230 |
+
"rewards/accuracies": 0.44999998807907104,
|
| 231 |
+
"rewards/chosen": -3.6857337951660156,
|
| 232 |
+
"rewards/margins": 1.9497820138931274,
|
| 233 |
+
"rewards/rejected": -5.6355156898498535,
|
| 234 |
+
"step": 70
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.6,
|
| 238 |
+
"grad_norm": 8.166132926940918,
|
| 239 |
+
"learning_rate": 0.0005,
|
| 240 |
+
"logits/chosen": -28.095006942749023,
|
| 241 |
+
"logits/rejected": -27.97015380859375,
|
| 242 |
+
"logps/chosen": -94.32670593261719,
|
| 243 |
+
"logps/rejected": -136.370849609375,
|
| 244 |
+
"loss": 1.4148,
|
| 245 |
+
"rewards/accuracies": 0.2750000059604645,
|
| 246 |
+
"rewards/chosen": -4.648624420166016,
|
| 247 |
+
"rewards/margins": 2.7192163467407227,
|
| 248 |
+
"rewards/rejected": -7.367840766906738,
|
| 249 |
+
"step": 75
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.64,
|
| 253 |
+
"grad_norm": 27.16705894470215,
|
| 254 |
+
"learning_rate": 0.0005333333333333334,
|
| 255 |
+
"logits/chosen": -29.13075828552246,
|
| 256 |
+
"logits/rejected": -28.229665756225586,
|
| 257 |
+
"logps/chosen": -97.07182312011719,
|
| 258 |
+
"logps/rejected": -96.37870788574219,
|
| 259 |
+
"loss": 2.0399,
|
| 260 |
+
"rewards/accuracies": 0.25,
|
| 261 |
+
"rewards/chosen": -5.250546455383301,
|
| 262 |
+
"rewards/margins": -0.06968679279088974,
|
| 263 |
+
"rewards/rejected": -5.180859565734863,
|
| 264 |
+
"step": 80
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.68,
|
| 268 |
+
"grad_norm": 5.628029823303223,
|
| 269 |
+
"learning_rate": 0.0005666666666666667,
|
| 270 |
+
"logits/chosen": -29.270288467407227,
|
| 271 |
+
"logits/rejected": -29.520370483398438,
|
| 272 |
+
"logps/chosen": -133.90028381347656,
|
| 273 |
+
"logps/rejected": -134.59854125976562,
|
| 274 |
+
"loss": 2.4778,
|
| 275 |
+
"rewards/accuracies": 0.4000000059604645,
|
| 276 |
+
"rewards/chosen": -6.5320563316345215,
|
| 277 |
+
"rewards/margins": 0.850067138671875,
|
| 278 |
+
"rewards/rejected": -7.382122993469238,
|
| 279 |
+
"step": 85
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.72,
|
| 283 |
+
"grad_norm": 11.795543670654297,
|
| 284 |
+
"learning_rate": 0.0006,
|
| 285 |
+
"logits/chosen": -22.93393898010254,
|
| 286 |
+
"logits/rejected": -22.469470977783203,
|
| 287 |
+
"logps/chosen": -121.58201599121094,
|
| 288 |
+
"logps/rejected": -107.5808334350586,
|
| 289 |
+
"loss": 3.0408,
|
| 290 |
+
"rewards/accuracies": 0.25,
|
| 291 |
+
"rewards/chosen": -7.048091888427734,
|
| 292 |
+
"rewards/margins": -1.3665062189102173,
|
| 293 |
+
"rewards/rejected": -5.681586265563965,
|
| 294 |
+
"step": 90
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 0.76,
|
| 298 |
+
"grad_norm": 4.0612053871154785,
|
| 299 |
+
"learning_rate": 0.0006333333333333333,
|
| 300 |
+
"logits/chosen": -12.232336044311523,
|
| 301 |
+
"logits/rejected": -12.792488098144531,
|
| 302 |
+
"logps/chosen": -138.4469451904297,
|
| 303 |
+
"logps/rejected": -137.1990203857422,
|
| 304 |
+
"loss": 2.6066,
|
| 305 |
+
"rewards/accuracies": 0.22499999403953552,
|
| 306 |
+
"rewards/chosen": -7.793431758880615,
|
| 307 |
+
"rewards/margins": -0.2190052568912506,
|
| 308 |
+
"rewards/rejected": -7.574427127838135,
|
| 309 |
+
"step": 95
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.8,
|
| 313 |
+
"grad_norm": 5.953216075897217,
|
| 314 |
+
"learning_rate": 0.0006666666666666666,
|
| 315 |
+
"logits/chosen": -19.619407653808594,
|
| 316 |
+
"logits/rejected": -19.87697410583496,
|
| 317 |
+
"logps/chosen": -106.26551818847656,
|
| 318 |
+
"logps/rejected": -144.4300537109375,
|
| 319 |
+
"loss": 2.2971,
|
| 320 |
+
"rewards/accuracies": 0.32499998807907104,
|
| 321 |
+
"rewards/chosen": -5.7664923667907715,
|
| 322 |
+
"rewards/margins": 1.8955154418945312,
|
| 323 |
+
"rewards/rejected": -7.662007808685303,
|
| 324 |
+
"step": 100
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.84,
|
| 328 |
+
"grad_norm": 7.588869094848633,
|
| 329 |
+
"learning_rate": 0.0007,
|
| 330 |
+
"logits/chosen": -9.69135856628418,
|
| 331 |
+
"logits/rejected": -10.370521545410156,
|
| 332 |
+
"logps/chosen": -125.16754150390625,
|
| 333 |
+
"logps/rejected": -176.91824340820312,
|
| 334 |
+
"loss": 1.3448,
|
| 335 |
+
"rewards/accuracies": 0.574999988079071,
|
| 336 |
+
"rewards/chosen": -6.616153717041016,
|
| 337 |
+
"rewards/margins": 3.5735087394714355,
|
| 338 |
+
"rewards/rejected": -10.189661979675293,
|
| 339 |
+
"step": 105
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.88,
|
| 343 |
+
"grad_norm": 0.3584274649620056,
|
| 344 |
+
"learning_rate": 0.0007333333333333333,
|
| 345 |
+
"logits/chosen": -22.10556411743164,
|
| 346 |
+
"logits/rejected": -21.52859878540039,
|
| 347 |
+
"logps/chosen": -100.20098876953125,
|
| 348 |
+
"logps/rejected": -173.8192901611328,
|
| 349 |
+
"loss": 1.2837,
|
| 350 |
+
"rewards/accuracies": 0.4000000059604645,
|
| 351 |
+
"rewards/chosen": -5.558084487915039,
|
| 352 |
+
"rewards/margins": 4.353057384490967,
|
| 353 |
+
"rewards/rejected": -9.911142349243164,
|
| 354 |
+
"step": 110
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.92,
|
| 358 |
+
"grad_norm": 0.5086020827293396,
|
| 359 |
+
"learning_rate": 0.0007666666666666667,
|
| 360 |
+
"logits/chosen": -12.327280044555664,
|
| 361 |
+
"logits/rejected": -9.589296340942383,
|
| 362 |
+
"logps/chosen": -115.45463562011719,
|
| 363 |
+
"logps/rejected": -167.5521697998047,
|
| 364 |
+
"loss": 1.0302,
|
| 365 |
+
"rewards/accuracies": 0.4000000059604645,
|
| 366 |
+
"rewards/chosen": -5.947112083435059,
|
| 367 |
+
"rewards/margins": 3.324498414993286,
|
| 368 |
+
"rewards/rejected": -9.271611213684082,
|
| 369 |
+
"step": 115
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.96,
|
| 373 |
+
"grad_norm": 4.804110050201416,
|
| 374 |
+
"learning_rate": 0.0008,
|
| 375 |
+
"logits/chosen": -9.970030784606934,
|
| 376 |
+
"logits/rejected": -9.584561347961426,
|
| 377 |
+
"logps/chosen": -101.1465835571289,
|
| 378 |
+
"logps/rejected": -150.82936096191406,
|
| 379 |
+
"loss": 1.4967,
|
| 380 |
+
"rewards/accuracies": 0.42500001192092896,
|
| 381 |
+
"rewards/chosen": -4.991378307342529,
|
| 382 |
+
"rewards/margins": 3.096414089202881,
|
| 383 |
+
"rewards/rejected": -8.087793350219727,
|
| 384 |
+
"step": 120
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.0,
|
| 388 |
+
"grad_norm": 2.8026299476623535,
|
| 389 |
+
"learning_rate": 0.0008333333333333334,
|
| 390 |
+
"logits/chosen": -9.677010536193848,
|
| 391 |
+
"logits/rejected": -9.15707015991211,
|
| 392 |
+
"logps/chosen": -93.2671127319336,
|
| 393 |
+
"logps/rejected": -111.13774108886719,
|
| 394 |
+
"loss": 1.4645,
|
| 395 |
+
"rewards/accuracies": 0.2750000059604645,
|
| 396 |
+
"rewards/chosen": -4.593766689300537,
|
| 397 |
+
"rewards/margins": 1.20462965965271,
|
| 398 |
+
"rewards/rejected": -5.798396110534668,
|
| 399 |
+
"step": 125
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.04,
|
| 403 |
+
"grad_norm": 2.7500274181365967,
|
| 404 |
+
"learning_rate": 0.0008666666666666667,
|
| 405 |
+
"logits/chosen": -16.292402267456055,
|
| 406 |
+
"logits/rejected": -15.907893180847168,
|
| 407 |
+
"logps/chosen": -130.10824584960938,
|
| 408 |
+
"logps/rejected": -147.6048583984375,
|
| 409 |
+
"loss": 1.5781,
|
| 410 |
+
"rewards/accuracies": 0.42500001192092896,
|
| 411 |
+
"rewards/chosen": -6.216085910797119,
|
| 412 |
+
"rewards/margins": 1.6796619892120361,
|
| 413 |
+
"rewards/rejected": -7.895747184753418,
|
| 414 |
+
"step": 130
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.08,
|
| 418 |
+
"grad_norm": 0.547558069229126,
|
| 419 |
+
"learning_rate": 0.0009000000000000001,
|
| 420 |
+
"logits/chosen": -26.10896873474121,
|
| 421 |
+
"logits/rejected": -24.641843795776367,
|
| 422 |
+
"logps/chosen": -100.40058898925781,
|
| 423 |
+
"logps/rejected": -135.31849670410156,
|
| 424 |
+
"loss": 1.8739,
|
| 425 |
+
"rewards/accuracies": 0.2750000059604645,
|
| 426 |
+
"rewards/chosen": -5.234818935394287,
|
| 427 |
+
"rewards/margins": 1.811771035194397,
|
| 428 |
+
"rewards/rejected": -7.0465898513793945,
|
| 429 |
+
"step": 135
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.12,
|
| 433 |
+
"grad_norm": 6.6305060386657715,
|
| 434 |
+
"learning_rate": 0.0009333333333333333,
|
| 435 |
+
"logits/chosen": -22.924015045166016,
|
| 436 |
+
"logits/rejected": -23.781505584716797,
|
| 437 |
+
"logps/chosen": -103.91690826416016,
|
| 438 |
+
"logps/rejected": -146.54783630371094,
|
| 439 |
+
"loss": 1.2655,
|
| 440 |
+
"rewards/accuracies": 0.3499999940395355,
|
| 441 |
+
"rewards/chosen": -5.9320454597473145,
|
| 442 |
+
"rewards/margins": 2.722485303878784,
|
| 443 |
+
"rewards/rejected": -8.65453052520752,
|
| 444 |
+
"step": 140
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.16,
|
| 448 |
+
"grad_norm": 15.08708381652832,
|
| 449 |
+
"learning_rate": 0.0009666666666666667,
|
| 450 |
+
"logits/chosen": -30.997777938842773,
|
| 451 |
+
"logits/rejected": -30.871929168701172,
|
| 452 |
+
"logps/chosen": -122.99382019042969,
|
| 453 |
+
"logps/rejected": -152.38113403320312,
|
| 454 |
+
"loss": 2.6552,
|
| 455 |
+
"rewards/accuracies": 0.42500001192092896,
|
| 456 |
+
"rewards/chosen": -6.272470474243164,
|
| 457 |
+
"rewards/margins": 2.4310927391052246,
|
| 458 |
+
"rewards/rejected": -8.70356273651123,
|
| 459 |
+
"step": 145
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.2,
|
| 463 |
+
"grad_norm": 5.697259426116943,
|
| 464 |
+
"learning_rate": 0.001,
|
| 465 |
+
"logits/chosen": -25.824504852294922,
|
| 466 |
+
"logits/rejected": -26.929336547851562,
|
| 467 |
+
"logps/chosen": -145.86514282226562,
|
| 468 |
+
"logps/rejected": -185.159912109375,
|
| 469 |
+
"loss": 1.9388,
|
| 470 |
+
"rewards/accuracies": 0.3499999940395355,
|
| 471 |
+
"rewards/chosen": -9.254964828491211,
|
| 472 |
+
"rewards/margins": 2.9817638397216797,
|
| 473 |
+
"rewards/rejected": -12.236729621887207,
|
| 474 |
+
"step": 150
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 1.24,
|
| 478 |
+
"grad_norm": 4.844988822937012,
|
| 479 |
+
"learning_rate": 0.0009000000000000001,
|
| 480 |
+
"logits/chosen": -38.678306579589844,
|
| 481 |
+
"logits/rejected": -42.88629913330078,
|
| 482 |
+
"logps/chosen": -144.60391235351562,
|
| 483 |
+
"logps/rejected": -180.99795532226562,
|
| 484 |
+
"loss": 2.4798,
|
| 485 |
+
"rewards/accuracies": 0.3499999940395355,
|
| 486 |
+
"rewards/chosen": -8.959487915039062,
|
| 487 |
+
"rewards/margins": 2.568267583847046,
|
| 488 |
+
"rewards/rejected": -11.527753829956055,
|
| 489 |
+
"step": 155
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.28,
|
| 493 |
+
"grad_norm": 10.417009353637695,
|
| 494 |
+
"learning_rate": 0.0008,
|
| 495 |
+
"logits/chosen": -40.77358627319336,
|
| 496 |
+
"logits/rejected": -40.55320739746094,
|
| 497 |
+
"logps/chosen": -91.55918884277344,
|
| 498 |
+
"logps/rejected": -91.25990295410156,
|
| 499 |
+
"loss": 1.9074,
|
| 500 |
+
"rewards/accuracies": 0.30000001192092896,
|
| 501 |
+
"rewards/chosen": -5.741996765136719,
|
| 502 |
+
"rewards/margins": 0.21963921189308167,
|
| 503 |
+
"rewards/rejected": -5.961636066436768,
|
| 504 |
+
"step": 160
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 1.32,
|
| 508 |
+
"grad_norm": 5.1186203956604,
|
| 509 |
+
"learning_rate": 0.0007,
|
| 510 |
+
"logits/chosen": -32.771751403808594,
|
| 511 |
+
"logits/rejected": -33.01805877685547,
|
| 512 |
+
"logps/chosen": -125.19621276855469,
|
| 513 |
+
"logps/rejected": -192.7315216064453,
|
| 514 |
+
"loss": 0.9249,
|
| 515 |
+
"rewards/accuracies": 0.550000011920929,
|
| 516 |
+
"rewards/chosen": -7.222474575042725,
|
| 517 |
+
"rewards/margins": 4.570401668548584,
|
| 518 |
+
"rewards/rejected": -11.792876243591309,
|
| 519 |
+
"step": 165
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.3599999999999999,
|
| 523 |
+
"grad_norm": 5.619666576385498,
|
| 524 |
+
"learning_rate": 0.0006,
|
| 525 |
+
"logits/chosen": -34.66514205932617,
|
| 526 |
+
"logits/rejected": -35.343589782714844,
|
| 527 |
+
"logps/chosen": -127.21083068847656,
|
| 528 |
+
"logps/rejected": -146.67799377441406,
|
| 529 |
+
"loss": 3.1172,
|
| 530 |
+
"rewards/accuracies": 0.32499998807907104,
|
| 531 |
+
"rewards/chosen": -7.435975551605225,
|
| 532 |
+
"rewards/margins": 1.6686077117919922,
|
| 533 |
+
"rewards/rejected": -9.104582786560059,
|
| 534 |
+
"step": 170
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.4,
|
| 538 |
+
"grad_norm": 11.079388618469238,
|
| 539 |
+
"learning_rate": 0.0005,
|
| 540 |
+
"logits/chosen": -29.087398529052734,
|
| 541 |
+
"logits/rejected": -28.334430694580078,
|
| 542 |
+
"logps/chosen": -137.70327758789062,
|
| 543 |
+
"logps/rejected": -148.99215698242188,
|
| 544 |
+
"loss": 3.1009,
|
| 545 |
+
"rewards/accuracies": 0.3499999940395355,
|
| 546 |
+
"rewards/chosen": -7.25787878036499,
|
| 547 |
+
"rewards/margins": 1.233079195022583,
|
| 548 |
+
"rewards/rejected": -8.490957260131836,
|
| 549 |
+
"step": 175
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.44,
|
| 553 |
+
"grad_norm": 9.70936393737793,
|
| 554 |
+
"learning_rate": 0.0004,
|
| 555 |
+
"logits/chosen": -24.516380310058594,
|
| 556 |
+
"logits/rejected": -25.09756851196289,
|
| 557 |
+
"logps/chosen": -119.39337158203125,
|
| 558 |
+
"logps/rejected": -132.81414794921875,
|
| 559 |
+
"loss": 1.151,
|
| 560 |
+
"rewards/accuracies": 0.375,
|
| 561 |
+
"rewards/chosen": -6.041991233825684,
|
| 562 |
+
"rewards/margins": 1.1351903676986694,
|
| 563 |
+
"rewards/rejected": -7.177182197570801,
|
| 564 |
+
"step": 180
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.48,
|
| 568 |
+
"grad_norm": 5.812276363372803,
|
| 569 |
+
"learning_rate": 0.0003,
|
| 570 |
+
"logits/chosen": -24.220096588134766,
|
| 571 |
+
"logits/rejected": -23.560684204101562,
|
| 572 |
+
"logps/chosen": -113.18685150146484,
|
| 573 |
+
"logps/rejected": -157.8691864013672,
|
| 574 |
+
"loss": 1.077,
|
| 575 |
+
"rewards/accuracies": 0.42500001192092896,
|
| 576 |
+
"rewards/chosen": -5.617026329040527,
|
| 577 |
+
"rewards/margins": 2.29874587059021,
|
| 578 |
+
"rewards/rejected": -7.915771484375,
|
| 579 |
+
"step": 185
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.52,
|
| 583 |
+
"grad_norm": 5.467133045196533,
|
| 584 |
+
"learning_rate": 0.0002,
|
| 585 |
+
"logits/chosen": -24.861865997314453,
|
| 586 |
+
"logits/rejected": -24.208385467529297,
|
| 587 |
+
"logps/chosen": -104.0861587524414,
|
| 588 |
+
"logps/rejected": -138.99636840820312,
|
| 589 |
+
"loss": 1.8515,
|
| 590 |
+
"rewards/accuracies": 0.375,
|
| 591 |
+
"rewards/chosen": -5.6537370681762695,
|
| 592 |
+
"rewards/margins": 1.677701711654663,
|
| 593 |
+
"rewards/rejected": -7.331439018249512,
|
| 594 |
+
"step": 190
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 1.56,
|
| 598 |
+
"grad_norm": 8.390186309814453,
|
| 599 |
+
"learning_rate": 0.0001,
|
| 600 |
+
"logits/chosen": -23.849029541015625,
|
| 601 |
+
"logits/rejected": -24.4769287109375,
|
| 602 |
+
"logps/chosen": -125.6314697265625,
|
| 603 |
+
"logps/rejected": -112.47705078125,
|
| 604 |
+
"loss": 2.5746,
|
| 605 |
+
"rewards/accuracies": 0.20000000298023224,
|
| 606 |
+
"rewards/chosen": -6.49404239654541,
|
| 607 |
+
"rewards/margins": -0.7069278359413147,
|
| 608 |
+
"rewards/rejected": -5.78711462020874,
|
| 609 |
+
"step": 195
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.6,
|
| 613 |
+
"grad_norm": 2.539565086364746,
|
| 614 |
+
"learning_rate": 0.0,
|
| 615 |
+
"logits/chosen": -23.671512603759766,
|
| 616 |
+
"logits/rejected": -23.140186309814453,
|
| 617 |
+
"logps/chosen": -92.0727767944336,
|
| 618 |
+
"logps/rejected": -97.2618408203125,
|
| 619 |
+
"loss": 1.1731,
|
| 620 |
+
"rewards/accuracies": 0.4000000059604645,
|
| 621 |
+
"rewards/chosen": -4.057224750518799,
|
| 622 |
+
"rewards/margins": 1.1420440673828125,
|
| 623 |
+
"rewards/rejected": -5.199268341064453,
|
| 624 |
+
"step": 200
|
| 625 |
+
}
|
| 626 |
+
],
|
| 627 |
+
"logging_steps": 5,
|
| 628 |
+
"max_steps": 200,
|
| 629 |
+
"num_input_tokens_seen": 0,
|
| 630 |
+
"num_train_epochs": 2,
|
| 631 |
+
"save_steps": 500,
|
| 632 |
+
"stateful_callbacks": {
|
| 633 |
+
"TrainerControl": {
|
| 634 |
+
"args": {
|
| 635 |
+
"should_epoch_stop": false,
|
| 636 |
+
"should_evaluate": false,
|
| 637 |
+
"should_log": false,
|
| 638 |
+
"should_save": true,
|
| 639 |
+
"should_training_stop": true
|
| 640 |
+
},
|
| 641 |
+
"attributes": {}
|
| 642 |
+
}
|
| 643 |
+
},
|
| 644 |
+
"total_flos": 0.0,
|
| 645 |
+
"train_batch_size": 8,
|
| 646 |
+
"trial_name": null,
|
| 647 |
+
"trial_params": null
|
| 648 |
+
}
|
test/checkpoint-200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e02125d9dae6e2a3d7b5d10dfd923a5ad9028df90d33dd14becad15c1dada561
|
| 3 |
+
size 6008
|
test/checkpoint-200/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
test/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 768,
|
| 16 |
+
"n_head": 12,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 12,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": false,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.46.0",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50257
|
| 39 |
+
}
|
test/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.46.0"
|
| 6 |
+
}
|
test/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
test/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b9cd4ea6d70556b7f7a032265263517f7730718cfafa3e7b911b4010c650282
|
| 3 |
+
size 497774208
|
test/runs/Feb28_18-34-47_puffer/events.out.tfevents.1740767688.puffer.3556740.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bfef6a0a0ff61786735e7a35c5f4873eaa2bbe57e6076916cf78c9b4b4e1136e
|
| 3 |
+
size 34262
|
test/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
test/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
test/tokenizer_config.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
}
|
| 12 |
+
},
|
| 13 |
+
"bos_token": "<|endoftext|>",
|
| 14 |
+
"clean_up_tokenization_spaces": false,
|
| 15 |
+
"eos_token": "<|endoftext|>",
|
| 16 |
+
"model_max_length": 1024,
|
| 17 |
+
"pad_token": "<|endoftext|>",
|
| 18 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 19 |
+
"unk_token": "<|endoftext|>"
|
| 20 |
+
}
|
test/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e02125d9dae6e2a3d7b5d10dfd923a5ad9028df90d33dd14becad15c1dada561
|
| 3 |
+
size 6008
|
test/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|