Upload 14 files
Browse files- added_tokens.json +8 -0
- config.json +39 -0
- generation_config.json +9 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +67 -0
- tokenizer.json +0 -0
- tokenizer_config.json +80 -0
- trainer_state.json +394 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|end|>": 50260,
|
| 3 |
+
"<|message|>": 50259,
|
| 4 |
+
"<|pad|>": 50257,
|
| 5 |
+
"<|start|>": 50258,
|
| 6 |
+
"Assistant:": 50262,
|
| 7 |
+
"User:": 50261
|
| 8 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"dtype": "float16",
|
| 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 |
+
"pad_token_id": 50257,
|
| 21 |
+
"reorder_and_upcast_attn": false,
|
| 22 |
+
"resid_pdrop": 0.1,
|
| 23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 24 |
+
"scale_attn_weights": true,
|
| 25 |
+
"summary_activation": null,
|
| 26 |
+
"summary_first_dropout": 0.1,
|
| 27 |
+
"summary_proj_to_labels": true,
|
| 28 |
+
"summary_type": "cls_index",
|
| 29 |
+
"summary_use_proj": true,
|
| 30 |
+
"task_specific_params": {
|
| 31 |
+
"text-generation": {
|
| 32 |
+
"do_sample": true,
|
| 33 |
+
"max_length": 50
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"transformers_version": "4.56.0",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50263
|
| 39 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
50256
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 50257,
|
| 8 |
+
"transformers_version": "4.56.0"
|
| 9 |
+
}
|
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:d844df17e8429bb4607a77b5b9cd94852df7a4b20f220e22c7b0582ec486a8b1
|
| 3 |
+
size 248903728
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36c02045553692aafe032dd3dac80165260d26bffaf6fc1ef665f5c4437d8921
|
| 3 |
+
size 497904075
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab8105924bb340e63f21ce0b297cd3cbfd2cbf97d947489830c0a9bf130f3f84
|
| 3 |
+
size 1465
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<|start|>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"content": "<|message|>",
|
| 12 |
+
"lstrip": false,
|
| 13 |
+
"normalized": false,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"content": "<|end|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"content": "User:",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": false,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"content": "Assistant:",
|
| 33 |
+
"lstrip": false,
|
| 34 |
+
"normalized": false,
|
| 35 |
+
"rstrip": false,
|
| 36 |
+
"single_word": false
|
| 37 |
+
}
|
| 38 |
+
],
|
| 39 |
+
"bos_token": {
|
| 40 |
+
"content": "<|endoftext|>",
|
| 41 |
+
"lstrip": false,
|
| 42 |
+
"normalized": false,
|
| 43 |
+
"rstrip": false,
|
| 44 |
+
"single_word": false
|
| 45 |
+
},
|
| 46 |
+
"eos_token": {
|
| 47 |
+
"content": "<|endoftext|>",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": false,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false
|
| 52 |
+
},
|
| 53 |
+
"pad_token": {
|
| 54 |
+
"content": "<|pad|>",
|
| 55 |
+
"lstrip": false,
|
| 56 |
+
"normalized": false,
|
| 57 |
+
"rstrip": false,
|
| 58 |
+
"single_word": false
|
| 59 |
+
},
|
| 60 |
+
"unk_token": {
|
| 61 |
+
"content": "<|endoftext|>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false
|
| 66 |
+
}
|
| 67 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"50257": {
|
| 13 |
+
"content": "<|pad|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"50258": {
|
| 21 |
+
"content": "<|start|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"50259": {
|
| 29 |
+
"content": "<|message|>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50260": {
|
| 37 |
+
"content": "<|end|>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"50261": {
|
| 45 |
+
"content": "User:",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"50262": {
|
| 53 |
+
"content": "Assistant:",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
}
|
| 60 |
+
},
|
| 61 |
+
"additional_special_tokens": [
|
| 62 |
+
"<|start|>",
|
| 63 |
+
"<|message|>",
|
| 64 |
+
"<|end|>",
|
| 65 |
+
"User:",
|
| 66 |
+
"Assistant:"
|
| 67 |
+
],
|
| 68 |
+
"bos_token": "<|endoftext|>",
|
| 69 |
+
"clean_up_tokenization_spaces": false,
|
| 70 |
+
"eos_token": "<|endoftext|>",
|
| 71 |
+
"extra_special_tokens": {},
|
| 72 |
+
"max_length": 512,
|
| 73 |
+
"model_max_length": 1024,
|
| 74 |
+
"pad_token": "<|pad|>",
|
| 75 |
+
"stride": 0,
|
| 76 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 77 |
+
"truncation_side": "right",
|
| 78 |
+
"truncation_strategy": "longest_first",
|
| 79 |
+
"unk_token": "<|endoftext|>"
|
| 80 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,394 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.2004500605850787,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1156,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0010386013501817552,
|
| 14 |
+
"grad_norm": 12.9140625,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"logits/chosen": -52.2838020324707,
|
| 17 |
+
"logits/rejected": -51.07138442993164,
|
| 18 |
+
"logps/chosen": -633.7007446289062,
|
| 19 |
+
"logps/rejected": -612.3391723632812,
|
| 20 |
+
"loss": 0.6931,
|
| 21 |
+
"rewards/accuracies": 0.0,
|
| 22 |
+
"rewards/chosen": 0.0,
|
| 23 |
+
"rewards/margins": 0.0,
|
| 24 |
+
"rewards/rejected": 0.0,
|
| 25 |
+
"step": 1
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.05193006750908776,
|
| 29 |
+
"grad_norm": 12.765625,
|
| 30 |
+
"learning_rate": 2.45e-07,
|
| 31 |
+
"logits/chosen": -52.915077209472656,
|
| 32 |
+
"logits/rejected": -53.6125373840332,
|
| 33 |
+
"logps/chosen": -554.205810546875,
|
| 34 |
+
"logps/rejected": -538.5724487304688,
|
| 35 |
+
"loss": 0.6936,
|
| 36 |
+
"rewards/accuracies": 0.5048894286155701,
|
| 37 |
+
"rewards/chosen": 0.0006169599364511669,
|
| 38 |
+
"rewards/margins": 0.009548303671181202,
|
| 39 |
+
"rewards/rejected": -0.00893134344369173,
|
| 40 |
+
"step": 50
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.10386013501817552,
|
| 44 |
+
"grad_norm": 14.1640625,
|
| 45 |
+
"learning_rate": 4.95e-07,
|
| 46 |
+
"logits/chosen": -52.115623474121094,
|
| 47 |
+
"logits/rejected": -52.78030776977539,
|
| 48 |
+
"logps/chosen": -559.9273071289062,
|
| 49 |
+
"logps/rejected": -523.4210205078125,
|
| 50 |
+
"loss": 0.6975,
|
| 51 |
+
"rewards/accuracies": 0.5493749976158142,
|
| 52 |
+
"rewards/chosen": -0.17299240827560425,
|
| 53 |
+
"rewards/margins": 0.04206321761012077,
|
| 54 |
+
"rewards/rejected": -0.2150556445121765,
|
| 55 |
+
"step": 100
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.15579020252726328,
|
| 59 |
+
"grad_norm": 12.359375,
|
| 60 |
+
"learning_rate": 4.767992424242424e-07,
|
| 61 |
+
"logits/chosen": -48.59254837036133,
|
| 62 |
+
"logits/rejected": -49.37075424194336,
|
| 63 |
+
"logps/chosen": -560.6456298828125,
|
| 64 |
+
"logps/rejected": -539.8798217773438,
|
| 65 |
+
"loss": 0.7206,
|
| 66 |
+
"rewards/accuracies": 0.5406249761581421,
|
| 67 |
+
"rewards/chosen": -0.6166394352912903,
|
| 68 |
+
"rewards/margins": 0.06871042400598526,
|
| 69 |
+
"rewards/rejected": -0.6853498816490173,
|
| 70 |
+
"step": 150
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.20772027003635105,
|
| 74 |
+
"grad_norm": 12.8984375,
|
| 75 |
+
"learning_rate": 4.53125e-07,
|
| 76 |
+
"logits/chosen": -46.112483978271484,
|
| 77 |
+
"logits/rejected": -46.65090560913086,
|
| 78 |
+
"logps/chosen": -565.4613037109375,
|
| 79 |
+
"logps/rejected": -539.76904296875,
|
| 80 |
+
"loss": 0.7316,
|
| 81 |
+
"rewards/accuracies": 0.5508333444595337,
|
| 82 |
+
"rewards/chosen": -0.8333625197410583,
|
| 83 |
+
"rewards/margins": 0.10787658393383026,
|
| 84 |
+
"rewards/rejected": -0.9412391185760498,
|
| 85 |
+
"step": 200
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.2596503375454388,
|
| 89 |
+
"grad_norm": 10.1484375,
|
| 90 |
+
"learning_rate": 4.2945075757575753e-07,
|
| 91 |
+
"logits/chosen": -41.89940643310547,
|
| 92 |
+
"logits/rejected": -42.62925338745117,
|
| 93 |
+
"logps/chosen": -587.5531616210938,
|
| 94 |
+
"logps/rejected": -545.7774047851562,
|
| 95 |
+
"loss": 0.7504,
|
| 96 |
+
"rewards/accuracies": 0.5518749952316284,
|
| 97 |
+
"rewards/chosen": -0.9503225684165955,
|
| 98 |
+
"rewards/margins": 0.10996780544519424,
|
| 99 |
+
"rewards/rejected": -1.0602904558181763,
|
| 100 |
+
"step": 250
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.31158040505452655,
|
| 104 |
+
"grad_norm": 9.3203125,
|
| 105 |
+
"learning_rate": 4.057765151515151e-07,
|
| 106 |
+
"logits/chosen": -39.214813232421875,
|
| 107 |
+
"logits/rejected": -39.6192626953125,
|
| 108 |
+
"logps/chosen": -574.7434692382812,
|
| 109 |
+
"logps/rejected": -546.6008911132812,
|
| 110 |
+
"loss": 0.7548,
|
| 111 |
+
"rewards/accuracies": 0.5558333396911621,
|
| 112 |
+
"rewards/chosen": -1.0960866212844849,
|
| 113 |
+
"rewards/margins": 0.12064453214406967,
|
| 114 |
+
"rewards/rejected": -1.2167311906814575,
|
| 115 |
+
"step": 300
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.36351047256361435,
|
| 119 |
+
"grad_norm": 11.7265625,
|
| 120 |
+
"learning_rate": 3.821022727272727e-07,
|
| 121 |
+
"logits/chosen": -35.89487075805664,
|
| 122 |
+
"logits/rejected": -36.3900260925293,
|
| 123 |
+
"logps/chosen": -573.7588500976562,
|
| 124 |
+
"logps/rejected": -550.8773193359375,
|
| 125 |
+
"loss": 0.7503,
|
| 126 |
+
"rewards/accuracies": 0.5472916960716248,
|
| 127 |
+
"rewards/chosen": -1.1505017280578613,
|
| 128 |
+
"rewards/margins": 0.12173371762037277,
|
| 129 |
+
"rewards/rejected": -1.2722352743148804,
|
| 130 |
+
"step": 350
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.4154405400727021,
|
| 134 |
+
"grad_norm": 9.9375,
|
| 135 |
+
"learning_rate": 3.5842803030303023e-07,
|
| 136 |
+
"logits/chosen": -37.06876754760742,
|
| 137 |
+
"logits/rejected": -37.600608825683594,
|
| 138 |
+
"logps/chosen": -580.6837768554688,
|
| 139 |
+
"logps/rejected": -555.3170166015625,
|
| 140 |
+
"loss": 0.7642,
|
| 141 |
+
"rewards/accuracies": 0.5475000143051147,
|
| 142 |
+
"rewards/chosen": -1.1745213270187378,
|
| 143 |
+
"rewards/margins": 0.10050401836633682,
|
| 144 |
+
"rewards/rejected": -1.275025486946106,
|
| 145 |
+
"step": 400
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.46737060758178983,
|
| 149 |
+
"grad_norm": 8.65625,
|
| 150 |
+
"learning_rate": 3.3475378787878787e-07,
|
| 151 |
+
"logits/chosen": -36.836238861083984,
|
| 152 |
+
"logits/rejected": -37.89125061035156,
|
| 153 |
+
"logps/chosen": -579.6170043945312,
|
| 154 |
+
"logps/rejected": -537.1629028320312,
|
| 155 |
+
"loss": 0.7622,
|
| 156 |
+
"rewards/accuracies": 0.5516666769981384,
|
| 157 |
+
"rewards/chosen": -1.1362614631652832,
|
| 158 |
+
"rewards/margins": 0.10563336312770844,
|
| 159 |
+
"rewards/rejected": -1.2418948411941528,
|
| 160 |
+
"step": 450
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.5193006750908776,
|
| 164 |
+
"grad_norm": 10.015625,
|
| 165 |
+
"learning_rate": 3.1107954545454546e-07,
|
| 166 |
+
"logits/chosen": -36.01582717895508,
|
| 167 |
+
"logits/rejected": -36.86990737915039,
|
| 168 |
+
"logps/chosen": -593.49267578125,
|
| 169 |
+
"logps/rejected": -540.7957153320312,
|
| 170 |
+
"loss": 0.7327,
|
| 171 |
+
"rewards/accuracies": 0.56270831823349,
|
| 172 |
+
"rewards/chosen": -1.8055402040481567,
|
| 173 |
+
"rewards/margins": 0.20275571942329407,
|
| 174 |
+
"rewards/rejected": -2.008295774459839,
|
| 175 |
+
"step": 500
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.5712307425999654,
|
| 179 |
+
"grad_norm": 9.1953125,
|
| 180 |
+
"learning_rate": 2.8740530303030304e-07,
|
| 181 |
+
"logits/chosen": -36.058692932128906,
|
| 182 |
+
"logits/rejected": -36.4224853515625,
|
| 183 |
+
"logps/chosen": -579.0709228515625,
|
| 184 |
+
"logps/rejected": -536.5722045898438,
|
| 185 |
+
"loss": 0.7522,
|
| 186 |
+
"rewards/accuracies": 0.5602083206176758,
|
| 187 |
+
"rewards/chosen": -1.2903023958206177,
|
| 188 |
+
"rewards/margins": 0.14331503212451935,
|
| 189 |
+
"rewards/rejected": -1.4336175918579102,
|
| 190 |
+
"step": 550
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.6231608101090531,
|
| 194 |
+
"grad_norm": 9.7578125,
|
| 195 |
+
"learning_rate": 2.6373106060606057e-07,
|
| 196 |
+
"logits/chosen": -36.24333953857422,
|
| 197 |
+
"logits/rejected": -36.79505157470703,
|
| 198 |
+
"logps/chosen": -579.2219848632812,
|
| 199 |
+
"logps/rejected": -542.4710693359375,
|
| 200 |
+
"loss": 0.7608,
|
| 201 |
+
"rewards/accuracies": 0.5543749928474426,
|
| 202 |
+
"rewards/chosen": -1.2316912412643433,
|
| 203 |
+
"rewards/margins": 0.14943528175354004,
|
| 204 |
+
"rewards/rejected": -1.3811265230178833,
|
| 205 |
+
"step": 600
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.6750908776181409,
|
| 209 |
+
"grad_norm": 11.5703125,
|
| 210 |
+
"learning_rate": 2.4005681818181816e-07,
|
| 211 |
+
"logits/chosen": -35.15838623046875,
|
| 212 |
+
"logits/rejected": -35.5240364074707,
|
| 213 |
+
"logps/chosen": -571.8494873046875,
|
| 214 |
+
"logps/rejected": -537.1287841796875,
|
| 215 |
+
"loss": 0.7319,
|
| 216 |
+
"rewards/accuracies": 0.5710416436195374,
|
| 217 |
+
"rewards/chosen": -1.079771876335144,
|
| 218 |
+
"rewards/margins": 0.18481211364269257,
|
| 219 |
+
"rewards/rejected": -1.2645841836929321,
|
| 220 |
+
"step": 650
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.7270209451272287,
|
| 224 |
+
"grad_norm": 12.671875,
|
| 225 |
+
"learning_rate": 2.1638257575757574e-07,
|
| 226 |
+
"logits/chosen": -34.670799255371094,
|
| 227 |
+
"logits/rejected": -35.52939987182617,
|
| 228 |
+
"logps/chosen": -593.4468994140625,
|
| 229 |
+
"logps/rejected": -543.4580688476562,
|
| 230 |
+
"loss": 0.7386,
|
| 231 |
+
"rewards/accuracies": 0.5754166841506958,
|
| 232 |
+
"rewards/chosen": -1.2700467109680176,
|
| 233 |
+
"rewards/margins": 0.1889582872390747,
|
| 234 |
+
"rewards/rejected": -1.4590051174163818,
|
| 235 |
+
"step": 700
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.7789510126363164,
|
| 239 |
+
"grad_norm": 9.875,
|
| 240 |
+
"learning_rate": 1.9270833333333332e-07,
|
| 241 |
+
"logits/chosen": -34.24915313720703,
|
| 242 |
+
"logits/rejected": -34.75949478149414,
|
| 243 |
+
"logps/chosen": -576.6466674804688,
|
| 244 |
+
"logps/rejected": -540.9093017578125,
|
| 245 |
+
"loss": 0.7279,
|
| 246 |
+
"rewards/accuracies": 0.5727083086967468,
|
| 247 |
+
"rewards/chosen": -1.4491314888000488,
|
| 248 |
+
"rewards/margins": 0.19473887979984283,
|
| 249 |
+
"rewards/rejected": -1.64387047290802,
|
| 250 |
+
"step": 750
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.8308810801454042,
|
| 254 |
+
"grad_norm": 10.1328125,
|
| 255 |
+
"learning_rate": 1.690340909090909e-07,
|
| 256 |
+
"logits/chosen": -34.45772933959961,
|
| 257 |
+
"logits/rejected": -35.198944091796875,
|
| 258 |
+
"logps/chosen": -584.421142578125,
|
| 259 |
+
"logps/rejected": -536.7197265625,
|
| 260 |
+
"loss": 0.7172,
|
| 261 |
+
"rewards/accuracies": 0.5831249952316284,
|
| 262 |
+
"rewards/chosen": -1.2970749139785767,
|
| 263 |
+
"rewards/margins": 0.21611928939819336,
|
| 264 |
+
"rewards/rejected": -1.51319420337677,
|
| 265 |
+
"step": 800
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.882811147654492,
|
| 269 |
+
"grad_norm": 9.8125,
|
| 270 |
+
"learning_rate": 1.4535984848484847e-07,
|
| 271 |
+
"logits/chosen": -36.15963363647461,
|
| 272 |
+
"logits/rejected": -36.554447174072266,
|
| 273 |
+
"logps/chosen": -554.3092041015625,
|
| 274 |
+
"logps/rejected": -533.0109252929688,
|
| 275 |
+
"loss": 0.7264,
|
| 276 |
+
"rewards/accuracies": 0.5735416412353516,
|
| 277 |
+
"rewards/chosen": -1.2292083501815796,
|
| 278 |
+
"rewards/margins": 0.20282161235809326,
|
| 279 |
+
"rewards/rejected": -1.4320299625396729,
|
| 280 |
+
"step": 850
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.9347412151635797,
|
| 284 |
+
"grad_norm": 10.8203125,
|
| 285 |
+
"learning_rate": 1.2168560606060605e-07,
|
| 286 |
+
"logits/chosen": -34.217220306396484,
|
| 287 |
+
"logits/rejected": -34.753658294677734,
|
| 288 |
+
"logps/chosen": -571.6567993164062,
|
| 289 |
+
"logps/rejected": -546.5286254882812,
|
| 290 |
+
"loss": 0.7142,
|
| 291 |
+
"rewards/accuracies": 0.5860416889190674,
|
| 292 |
+
"rewards/chosen": -1.0470510721206665,
|
| 293 |
+
"rewards/margins": 0.22074727714061737,
|
| 294 |
+
"rewards/rejected": -1.2677983045578003,
|
| 295 |
+
"step": 900
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.9866712826726675,
|
| 299 |
+
"grad_norm": 9.1640625,
|
| 300 |
+
"learning_rate": 9.801136363636363e-08,
|
| 301 |
+
"logits/chosen": -33.653926849365234,
|
| 302 |
+
"logits/rejected": -34.451416015625,
|
| 303 |
+
"logps/chosen": -584.2894897460938,
|
| 304 |
+
"logps/rejected": -545.4525756835938,
|
| 305 |
+
"loss": 0.7221,
|
| 306 |
+
"rewards/accuracies": 0.5768749713897705,
|
| 307 |
+
"rewards/chosen": -1.1478312015533447,
|
| 308 |
+
"rewards/margins": 0.21039748191833496,
|
| 309 |
+
"rewards/rejected": -1.3582286834716797,
|
| 310 |
+
"step": 950
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.038428249956725,
|
| 314 |
+
"grad_norm": 4.6640625,
|
| 315 |
+
"learning_rate": 7.433712121212122e-08,
|
| 316 |
+
"logits/chosen": -34.1423225402832,
|
| 317 |
+
"logits/rejected": -34.61945343017578,
|
| 318 |
+
"logps/chosen": -576.2367553710938,
|
| 319 |
+
"logps/rejected": -549.8435668945312,
|
| 320 |
+
"loss": 0.4573,
|
| 321 |
+
"rewards/accuracies": 0.8028846383094788,
|
| 322 |
+
"rewards/chosen": -0.7547619938850403,
|
| 323 |
+
"rewards/margins": 1.0344306230545044,
|
| 324 |
+
"rewards/rejected": -1.7891926765441895,
|
| 325 |
+
"step": 1000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.0903583174658127,
|
| 329 |
+
"grad_norm": 5.54296875,
|
| 330 |
+
"learning_rate": 5.066287878787878e-08,
|
| 331 |
+
"logits/chosen": -35.22718811035156,
|
| 332 |
+
"logits/rejected": -35.88585662841797,
|
| 333 |
+
"logps/chosen": -575.0949096679688,
|
| 334 |
+
"logps/rejected": -546.9916381835938,
|
| 335 |
+
"loss": 0.3604,
|
| 336 |
+
"rewards/accuracies": 0.8833333253860474,
|
| 337 |
+
"rewards/chosen": -0.7632372975349426,
|
| 338 |
+
"rewards/margins": 1.3589388132095337,
|
| 339 |
+
"rewards/rejected": -2.1221764087677,
|
| 340 |
+
"step": 1050
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.1422883849749004,
|
| 344 |
+
"grad_norm": 6.796875,
|
| 345 |
+
"learning_rate": 2.6988636363636364e-08,
|
| 346 |
+
"logits/chosen": -35.756492614746094,
|
| 347 |
+
"logits/rejected": -36.60533142089844,
|
| 348 |
+
"logps/chosen": -573.737060546875,
|
| 349 |
+
"logps/rejected": -554.9136352539062,
|
| 350 |
+
"loss": 0.3588,
|
| 351 |
+
"rewards/accuracies": 0.8787500262260437,
|
| 352 |
+
"rewards/chosen": -0.839233934879303,
|
| 353 |
+
"rewards/margins": 1.390368938446045,
|
| 354 |
+
"rewards/rejected": -2.229602575302124,
|
| 355 |
+
"step": 1100
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 1.1942184524839883,
|
| 359 |
+
"grad_norm": 6.671875,
|
| 360 |
+
"learning_rate": 3.314393939393939e-09,
|
| 361 |
+
"logits/chosen": -35.812347412109375,
|
| 362 |
+
"logits/rejected": -36.6298942565918,
|
| 363 |
+
"logps/chosen": -566.4522094726562,
|
| 364 |
+
"logps/rejected": -543.6142578125,
|
| 365 |
+
"loss": 0.3581,
|
| 366 |
+
"rewards/accuracies": 0.8804166913032532,
|
| 367 |
+
"rewards/chosen": -0.9078888893127441,
|
| 368 |
+
"rewards/margins": 1.3726412057876587,
|
| 369 |
+
"rewards/rejected": -2.2805302143096924,
|
| 370 |
+
"step": 1150
|
| 371 |
+
}
|
| 372 |
+
],
|
| 373 |
+
"logging_steps": 50,
|
| 374 |
+
"max_steps": 1156,
|
| 375 |
+
"num_input_tokens_seen": 0,
|
| 376 |
+
"num_train_epochs": 2,
|
| 377 |
+
"save_steps": 900,
|
| 378 |
+
"stateful_callbacks": {
|
| 379 |
+
"TrainerControl": {
|
| 380 |
+
"args": {
|
| 381 |
+
"should_epoch_stop": false,
|
| 382 |
+
"should_evaluate": false,
|
| 383 |
+
"should_log": false,
|
| 384 |
+
"should_save": true,
|
| 385 |
+
"should_training_stop": true
|
| 386 |
+
},
|
| 387 |
+
"attributes": {}
|
| 388 |
+
}
|
| 389 |
+
},
|
| 390 |
+
"total_flos": 0.0,
|
| 391 |
+
"train_batch_size": 16,
|
| 392 |
+
"trial_name": null,
|
| 393 |
+
"trial_params": null
|
| 394 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c5b57a6ba09117026a683db3d86a977419374a389697d7ade2484fabb039977
|
| 3 |
+
size 6801
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|