INDEX

Explanations

references to locations or venues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ly

-0.19

arkan

-0.18

raq

-0.17

thew

-0.17

sse

-0.16

thin

-0.16

rades

-0.15

ishly

-0.15

sis

-0.15

rlen

-0.15

POSITIVE LOGITS

bos

0.37

HOLDER

0.32

-holder

0.30

holders

0.27

 holder

0.27

 where

0.26

holding

0.25

holder

0.25

Holder

0.24

lessness

0.24

Activations Density 0.061%

references to locations or venues

No Comments

No Known Activations