chchen commited on
Commit
98def46
·
verified ·
1 Parent(s): 1ecd0de

Training in progress, step 50

Browse files
adapter_config.json CHANGED
@@ -20,12 +20,12 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "down_proj",
24
- "gate_proj",
25
- "up_proj",
26
  "q_proj",
27
- "o_proj",
28
  "v_proj",
 
 
 
 
29
  "k_proj"
30
  ],
31
  "task_type": "CAUSAL_LM",
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
 
 
23
  "q_proj",
 
24
  "v_proj",
25
+ "down_proj",
26
+ "o_proj",
27
+ "up_proj",
28
+ "gate_proj",
29
  "k_proj"
30
  ],
31
  "task_type": "CAUSAL_LM",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3858e3f7e72ff77fc7ed02b660fa746aaea0646e9de9b9b7770635af08569315
3
  size 83945296
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:460398e833c2d1c3f9c3b818a30ddc07068a8171e6afbe04325f15879e1c8a85
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:dee9e454c8741be114a126b148d9bd9261cb548134cc1c09ca8dd88ab6ac3191
3
  size 5624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2b50dfac7a46a8ed5a719768c8c70daa4ae751fcceb2ee5e66b9fc90681780a
3
  size 5624