INDEX

Explanations

references to the act of reading or engagement with books

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

usat

-0.17

outil

-0.16

 æ¨

-0.16

 cycle

-0.15

ssf

-0.15

zem

-0.15

atif

-0.15

aho

-0.15

wid

-0.14

ledi

-0.14

POSITIVE LOGITS

aspers

0.19

ownik

0.16

igu

0.16

å¿ł

0.16

oki

0.15

As

0.15

iska

0.15

 ÐºÐ°Ð·

0.15

ilty

0.15

Byl

0.14

Activations Density 0.030%

references to the act of reading or engagement with books

No Comments

No Known Activations