INDEX

Explanations

mentions of "Cleveland" and variations of the name in context

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

eller

-0.18

lez

-0.16

 Wich

-0.15

uant

-0.14

ÑĢÐ°Ð¶

-0.14

eln

-0.14

ateurs

-0.14

ziej

-0.14

sch

-0.14

uml

-0.14

POSITIVE LOGITS

avage

0.23

cle

0.22

arence

0.20

Cle

0.19

ary

0.19

aver

0.18

op

0.18

mons

0.17

 Verg

0.17

aving

0.17

Activations Density 0.006%

mentions of "Cleveland" and variations of the name in context

No Comments

No Known Activations