Hugging Face's logo

distily
/
distily_profile_smollm

TensorBoard
Safetensors
Distily
llama
Generated from Trainer
Model card Files Metrics Community
lapp0 commited on
Commit
d6f28df
•
1 Parent(s): f774d99

Training in progress, step 5000

Browse files
Files changed (7) hide show
  1. config.json +1 -2
  2. logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=bs4_NO_liger_baseline, student_model_use_liger=False/events.out.tfevents.1726140851.1c1a426a2fee +3 -0
  3. logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=bs4_NO_liger_baseline, student_model_use_liger=True/events.out.tfevents.1726140475.1c1a426a2fee +3 -0
  4. logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=logits_only_bs4_liger, student_model_use_liger=True/completed.flag +0 -0
  5. logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=logits_only_bs4_liger, student_model_use_liger=True/events.out.tfevents.1726140101.1c1a426a2fee +2 -2
  6. model.safetensors +1 -1
  7. training_args.bin +2 -2
config.json CHANGED
@@ -7,7 +7,6 @@
7
  "attention_dropout": 0.0,
8
  "bos_token_id": 0,
9
  "eos_token_id": 0,
10
- "head_dim": 64,
11
  "hidden_act": "silu",
12
  "hidden_size": 576,
13
  "initializer_range": 0.02,
@@ -24,7 +23,7 @@
24
  "rope_theta": 10000.0,
25
  "tie_word_embeddings": true,
26
  "torch_dtype": "bfloat16",
27
- "transformers_version": "4.45.0.dev0",
28
  "use_cache": false,
29
  "vocab_size": 49152
30
  }
 
7
  "attention_dropout": 0.0,
8
  "bos_token_id": 0,
9
  "eos_token_id": 0,
 
10
  "hidden_act": "silu",
11
  "hidden_size": 576,
12
  "initializer_range": 0.02,
 
23
  "rope_theta": 10000.0,
24
  "tie_word_embeddings": true,
25
  "torch_dtype": "bfloat16",
26
+ "transformers_version": "4.44.2",
27
  "use_cache": false,
28
  "vocab_size": 49152
29
  }
logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=bs4_NO_liger_baseline, student_model_use_liger=False/events.out.tfevents.1726140851.1c1a426a2fee ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:376dd388ce2d98efce1e7abc3d76ff4c0587beabc647bdd8db714eda54e09c15
3
+ size 140519
logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=bs4_NO_liger_baseline, student_model_use_liger=True/events.out.tfevents.1726140475.1c1a426a2fee ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4325780a04a8c0815d05d6ec86801da2f1f225adf1dbd54e32d3635fb16decb2
3
+ size 20602
logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=logits_only_bs4_liger, student_model_use_liger=True/completed.flag ADDED
File without changes
logs/attn_weight=0.0, per_device_train_batch_size=4, run_name=logits_only_bs4_liger, student_model_use_liger=True/events.out.tfevents.1726140101.1c1a426a2fee CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d4be5e938388209e7ad5d49e063c3176e7d9d8d2a79479491ecb04af54c3d03c
3
- size 253
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:308dd9085aea786a80f3ce4745552a7a1ca411f9ab3a9069aee2fa6883f9c32f
3
+ size 529
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4daca763ebd90dc016f63434e4ebcc90c001011bb0721131b341d053ace96c8
3
  size 162842416
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56ec77a86d23f1e5ef6cf8170b575708bbfbb7ac6d93c849226bf57c5e1d64d4
3
  size 162842416
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:738e9326fbb30717baf30f64578731beac353bb8abd9108ce8783db4565fa56e
3
- size 5368
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:555ceb45b35190b72c7d1a0f76611e172bacea5f42f7a7ad17b8bf6a95802281
3
+ size 5432