Downloads · 30 days
127
6% of all-time downloads
RichardErkhov/diabolic6045_-_open-llama-3.2-1B-Instruct-gguf
diabolic6045_-_open-llama-3.2-1B-Instruct-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
127
6% of all-time downloads
All-time downloads
2.2K
Public
Repo size
15.9 GB
Likes
0
Public
Click a slice to open those files.
.gguf18.8 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
open-llama-3.2-1B-Instruct - GGUF
| Name | Quant method | Size |
|---|---|---|
| open-llama-3.2-1B-Instruct.Q2_K.gguf | Q2_K | 0.62GB |
| open-llama-3.2-1B-Instruct.Q3_K_S.gguf | Q3_K_S | 0.7GB |
| open-llama-3.2-1B-Instruct.Q3_K.gguf | Q3_K | 0.75GB |
| open-llama-3.2-1B-Instruct.Q3_K_M.gguf | Q3_K_M | 0.75GB |
| open-llama-3.2-1B-Instruct.Q3_K_L.gguf | Q3_K_L | 0.79GB |
| open-llama-3.2-1B-Instruct.IQ4_XS.gguf | IQ4_XS | 0.83GB |
| open-llama-3.2-1B-Instruct.Q4_0.gguf | Q4_0 | 0.86GB |
| open-llama-3.2-1B-Instruct.IQ4_NL.gguf | IQ4_NL | 0.86GB |
| open-llama-3.2-1B-Instruct.Q4_K_S.gguf | Q4_K_S | 0.86GB |
| open-llama-3.2-1B-Instruct.Q4_K.gguf | Q4_K | 0.89GB |
| open-llama-3.2-1B-Instruct.Q4_K_M.gguf | Q4_K_M | 0.89GB |
| open-llama-3.2-1B-Instruct.Q4_1.gguf | Q4_1 | 0.93GB |
| open-llama-3.2-1B-Instruct.Q5_0.gguf | Q5_0 | 1.0GB |
| open-llama-3.2-1B-Instruct.Q5_K_S.gguf | Q5_K_S | 1.0GB |
| open-llama-3.2-1B-Instruct.Q5_K.gguf | Q5_K | 1.02GB |
| open-llama-3.2-1B-Instruct.Q5_K_M.gguf | Q5_K_M | 1.02GB |
| open-llama-3.2-1B-Instruct.Q5_1.gguf | Q5_1 | 1.07GB |
| open-llama-3.2-1B-Instruct.Q6_K.gguf | Q6_K | 1.15GB |
| open-llama-3.2-1B-Instruct.Q8_0.gguf | Q8_0 | 1.49GB |
library_name: transformers license: llama3.2 base_model: meta-llama/Llama-3.2-1B-Instruct tags:
The following hyperparameters were used during training:
| Groups | Version | Filter | n-shot | Metric | Value | Stderr | ||
|---|---|---|---|---|---|---|---|---|
| mmlu | 2 | none | acc | ↑ | 0.3632 | ± | 0.0040 | |
| - humanities | 2 | none | acc | ↑ | 0.3411 | ± | 0.0068 | |
| - other | 2 | none | acc | ↑ | 0.4078 | ± | 0.0087 | |
| - social sciences | 2 | none | acc | ↑ | 0.3997 | ± | 0.0087 | |
| - stem | 2 | none | acc | ↑ | 0.3165 | ± | 0.0082 |
axolotl version: 0.4.1
base_model: meta-llama/Llama-3.2-1B-Instruct
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: diabolic6045/OpenHermes-2.5_alpaca_10
type: alpaca
dataset_prepared_path: last_run_prepared
val_set_size: 0
output_dir: ./outputs/out
hub_model_id: diabolic6045/open-llama-Instruct
hf_use_auth_token: true
sequence_len: 1024
sample_packing: true
pad_to_sequence_len: true
wandb_project: open-llama
wandb_entity:
wandb_watch: all
wandb_name: open-llama
wandb_log_model:
gradient_accumulation_steps: 1
micro_batch_size: 2
num_epochs: 1
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 2e-5
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
gradient_checkpointing_kwargs:
use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: false
warmup_steps: 10
evals_per_epoch: 2
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
pad_token: <|end_of_text|>
</details><br>