Downloads · 30 days
5
3% of all-time downloads
shisa-ai/ablation-32-rafbestseq.reasoning-shisa-v2-llama-3.1-8b
ablation-32-rafbestseq.reasoning-shisa-v2-llama-3.1-8b is a text generation model from shisa-ai. Use it when you need the model to write or continue text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
Downloads · 30 days
5
3% of all-time downloads
All-time downloads
176
Public
Parameters
8B
16.1 GB on disk
Likes
1
Public
Click a slice to open those files.
.safetensors16.1 GB · 100%
From the Hugging Face model README
axolotl version: 0.6.0
# We train the exact same as 02-llama31 but we swap gradient accum and micro bs (should be same but faster?)
base_model: /fsx2/outputs/ablation-26-reasoning.inputs-shisa-v2-llama-3.1-8b-lr8e6
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer
load_in_8bit: false
load_in_4bit: false
strict: false
# User Liger
plugins:
- axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_fused_linear_cross_entropy: true
chat_template: llama3
datasets:
- path: /fsx/ubuntu/meti/data/shisa-v1-bestseq-reannoatated-filtered
# type: sharegpt deprecated
type: chat_template
field_messages: conversations
message_field_role: from
message_field_content: value
shuffle_merged_datasets: false
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./outputs/ablation-32-rafbestseq.reasoning-shisa-v2-llama-3.1-8b
sequence_len: 8192
sample_packing: true
pad_to_sequence_len: true
# marginal difference
neftune_noise_alpha: 5
use_wandb: true
wandb_project: shisa-v2
wandb_entity: augmxnt
wandb_name: ablation-32-rafbestseq.reasoning-shisa-v2-llama-3.1-8b
gradient_accumulation_steps: 2
micro_batch_size: 4
num_epochs: 3
optimizer: paged_adamw_8bit
lr_scheduler: linear
learning_rate: 8e-6
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: true
warmup_ratio: 0.05
evals_per_epoch: 2
eval_table_size:
saves_per_epoch: 0
save_total_limit: 1 # Only store a single checkpoint
debug:
deepspeed: zero3_bf16.json
weight_decay: 1e-4
fsdp:
fsdp_config:
special_tokens:
pad_token: <|end_of_text|>
</details><br>
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
More information needed
More information needed
More information needed
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.046 | 0.0019 | 1 | 1.0658 |
| 0.6637 | 0.5005 | 257 | 0.4651 |
| 0.6707 | 1.0 | 514 | 0.4423 |
| 0.5639 | 1.5005 | 771 | 0.4403 |
| 0.5246 | 2.0 | 1028 | 0.4353 |
| 0.4838 | 2.5005 | 1285 | 0.4438 |