INDEX

Explanations

references to historical or fictional female heroes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ners

-0.18

peng

-0.16

olas

-0.16

gang

-0.15

ster

-0.15

ty

-0.15

mate

-0.14

ety

-0.14

ings

-0.14

à¸¡

-0.14

POSITIVE LOGITS

ines

0.24

ically

0.22

ics

0.20

anova

0.18

ine

0.18

ism

0.17

itics

0.16

ic

0.16

MES

0.16

Activations Density 0.015%

references to historical or fictional female heroes

No Comments

No Known Activations