INDEX

Explanations

references to gender and pet-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

aidu

-0.16

alsa

-0.15

oice

-0.15

oir

-0.15

uur

-0.15

osu

-0.15

akis

-0.15

erset

-0.14

ÑĥÐ³

-0.14

POSITIVE LOGITS

 alike

0.19

AMS

0.17

ams

0.15

alc

0.14

engo

0.14

apore

0.14

DDL

0.14

 separately

0.14

 norms

0.14

MCS

0.14

Activations Density 0.130%

references to gender and pet-related terms

No Comments

No Known Activations