INDEX

Explanations

references to "here" or navigation-related phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

xc

-0.15

Ð¾Ð»Ð¾Ñģ

-0.14

itaire

-0.14

lant

-0.14

laÄį

-0.14

elden

-0.14

 ÑĤÐ¾ÑĢ

-0.14

ionale

-0.14

 Nimbus

-0.14

EDIUM

-0.13

POSITIVE LOGITS

 Home

0.19

Home

0.18

kes

0.17

_HOME

0.16

 Stall

0.15

 heavily

0.14

 Skip

0.14

HOME

0.14

 home

0.14

ÏĢÎ·

0.14

Activations Density 0.003%

references to "here" or navigation-related phrases

No Comments

No Known Activations