Hugging Face's logo

gradientai
/
Llama-3-8B-Instruct-Gradient-1048k

Text Generation
Transformers
Safetensors
English
llama
meta
llama-3
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
27
forrest-gradient commited on
Commit
51a5569
•
1 Parent(s): c155d16

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -2
README.md CHANGED
@@ -41,13 +41,12 @@ For training data, we generate long contexts by augmenting [SlimPajama](https://
41
  | Initialize From | LLaMA-3 7B| 65K | 262K | 524k |
42
  | Sequence Length 2^N | 16 | 18 | 19 | 20 |
43
  | RoPE theta | 15.3 M | 207.1 M | 1.06B | 2.80B |
44
- | Batch Size | 1 | 1 | 2 | 2 |
45
  | Gradient Accumulation Steps | 32 | 16 | 1 | 1 |
46
  | Steps | 30 | 24 | 50 | 50 |
47
  | Total Tokens | 62914560 | 100663296 | 419430400 | 838860800 |
48
  | Learning Rate | 2.00E-05 | 2.00E-05 | 2.00E-05 | 2.00E-05 |
49
  | # GPUs | 8 | 32 | 512 | 512 |
50
- | Ring parallelism | 1 | 1 | 8 | 8 |
51
  | GPU Type | NVIDIA L40S | NVIDIA L40S | NVIDIA L40S | NVIDIA L40S |
52
  | Minutes to Train (Wall)| 202 | 555 | 61 | 87 |
53
 
 
41
  | Initialize From | LLaMA-3 7B| 65K | 262K | 524k |
42
  | Sequence Length 2^N | 16 | 18 | 19 | 20 |
43
  | RoPE theta | 15.3 M | 207.1 M | 1.06B | 2.80B |
44
+ | Batch Size | 1 | 1 | 16 | 16 |
45
  | Gradient Accumulation Steps | 32 | 16 | 1 | 1 |
46
  | Steps | 30 | 24 | 50 | 50 |
47
  | Total Tokens | 62914560 | 100663296 | 419430400 | 838860800 |
48
  | Learning Rate | 2.00E-05 | 2.00E-05 | 2.00E-05 | 2.00E-05 |
49
  | # GPUs | 8 | 32 | 512 | 512 |
 
50
  | GPU Type | NVIDIA L40S | NVIDIA L40S | NVIDIA L40S | NVIDIA L40S |
51
  | Minutes to Train (Wall)| 202 | 555 | 61 | 87 |
52