INDEX

Explanations

detailed descriptions of narratives and their elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 paraph

-0.15

ç½

-0.15

_CPP

-0.15

/books

-0.15

ablo

-0.14

arrow

-0.14

xygen

-0.14

CCCCCC

-0.14

 retr

-0.14

_exempt

-0.13

POSITIVE LOGITS

 stories

0.52

 Stories

0.46

 short

0.44

 story

0.43

stories

0.43

Stories

0.43

short

0.39

 shorts

0.37

-short

0.37

 Short

0.37

Activations Density 0.132%

detailed descriptions of narratives and their elements

No Comments

No Known Activations