chchen commited on
Commit
e972d95
·
verified ·
1 Parent(s): eae5ee0

Training in progress, step 50

Browse files
adapter_config.json CHANGED
@@ -20,13 +20,13 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
 
 
23
  "gate_proj",
24
  "q_proj",
25
  "down_proj",
26
- "v_proj",
27
- "k_proj",
28
- "up_proj",
29
- "o_proj"
30
  ],
31
  "task_type": "CAUSAL_LM",
32
  "use_dora": false,
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
+ "up_proj",
24
+ "v_proj",
25
+ "o_proj",
26
  "gate_proj",
27
  "q_proj",
28
  "down_proj",
29
+ "k_proj"
 
 
 
30
  ],
31
  "task_type": "CAUSAL_LM",
32
  "use_dora": false,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fd77ef110e521eddf68c1128bd93f757a9936bb2a817e2eb6c1e7f269ab54ae5
3
  size 83945296
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9555d9e90f2bf26fcd1707a7ab8ca51997fe886175023857a6b0dbd2fe41fc18
3
  size 83945296
llama3_lora_sft.yaml CHANGED
@@ -32,7 +32,7 @@ load_best_model_at_end: true
32
  per_device_train_batch_size: 1
33
  gradient_accumulation_steps: 16
34
  learning_rate: 1.0e-4
35
- num_train_epochs: 10.0
36
  lr_scheduler_type: cosine
37
  warmup_ratio: 0.1
38
  bf16: true
 
32
  per_device_train_batch_size: 1
33
  gradient_accumulation_steps: 16
34
  learning_rate: 1.0e-4
35
+ num_train_epochs: 5.0
36
  lr_scheduler_type: cosine
37
  warmup_ratio: 0.1
38
  bf16: true
trainer_log.jsonl CHANGED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ee2111c336fe1f9f606527a331619a63c29b3a3002e835a449f878357da46758
3
  size 5624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db10fcb36c11fe6e3f77bd5560ff600fa59fe13096095c3a662e0904c23543d3
3
  size 5624