INDEX

Explanations

variations of the word "success."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

tery

-0.17

 Bair

-0.16

estruction

-0.15

íĭĢ

-0.15

#__

-0.15

atsby

-0.15

 eigen

-0.15

Ú¯ÛĮØ±

-0.15

tn

-0.14

.Quad

-0.14

POSITIVE LOGITS

ulent

0.36

umbing

0.34

umbs

0.31

umbed

0.30

umb

0.30

cess

0.25

esse

0.23

ubi

0.23

ulence

0.22

inct

0.22

Activations Density 0.003%

variations of the word "success."

No Comments

No Known Activations

variations of the word "success."

No Comments

No Known Activations