Hugging Face's logo

gshasiri
/
SmolLM3-Mid

Text Generation
Transformers
Safetensors
llama
Generated from Trainer
trl
sft
alignment-handbook
conversational
text-generation-inference
Model card Files
xet
 Community
gshasiri commited on
Commit
749f266
·
verified ·
1 Parent(s): 68cc3f3

Model save

Browse files
Files changed (4) hide show
  1. all_results.json +9 -0
  2. generation_config.json +1 -4
  3. train_results.json +9 -0
  4. trainer_state.json +0 -0
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "total_flos": 4.107393905038092e+20,
4
+ "train_loss": 0.4648131006222995,
5
+ "train_runtime": 276842.9058,
6
+ "train_samples": 4779894,
7
+ "train_samples_per_second": 7.781,
8
+ "train_steps_per_second": 0.041
9
+ }
generation_config.json CHANGED
@@ -2,10 +2,7 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 128000,
4
  "do_sample": true,
5
- "eos_token_id": [
6
- 128012,
7
- 128001
8
- ],
9
  "temperature": 0.6,
10
  "top_p": 0.9,
11
  "transformers_version": "4.57.1"
 
2
  "_from_model_config": true,
3
  "bos_token_id": 128000,
4
  "do_sample": true,
5
+ "eos_token_id": 128012,
 
 
 
6
  "temperature": 0.6,
7
  "top_p": 0.9,
8
  "transformers_version": "4.57.1"
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "total_flos": 4.107393905038092e+20,
4
+ "train_loss": 0.4648131006222995,
5
+ "train_runtime": 276842.9058,
6
+ "train_samples": 4779894,
7
+ "train_samples_per_second": 7.781,
8
+ "train_steps_per_second": 0.041
9
+ }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff