INDEX

Explanations

references to geographic locations and places, particularly islands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

und

-0.15

odus

-0.15

 cork

-0.15

anje

-0.14

lox

-0.14

gon

-0.14

 elast

-0.14

unto

-0.14

sol

-0.13

ibr

-0.13

POSITIVE LOGITS

Tah

0.25

 Moore

0.21

Hao

0.21

tah

0.20

 Society

0.18

Fak

0.18

tah

0.17

 French

0.17

opa

0.17

Activations Density 0.007%

references to geographic locations and places, particularly islands

No Comments

No Known Activations