INDEX

Explanations

references to a person named Kate

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 mature

-0.15

ÅĤaw

-0.15

rado

-0.15

umd

-0.15

IFT

-0.15

thern

-0.14

weigh

-0.14

 revert

-0.14

theast

-0.14

ibaba

-0.14

POSITIVE LOGITS

ÅĻ

0.27

0.24

 Middleton

0.19

 Moss

0.18

 Mara

0.17

 middle

0.16

ych

0.16

 Wins

0.15

ky

0.14

 Hudson

0.14

Activations Density 0.006%

references to a person named Kate

No Comments

No Known Activations