INDEX

Explanations

variations of the word "board" in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

mland

-0.15

ebra

-0.14

avs

-0.14

 Variable

-0.14

ocities

-0.13

ason

-0.13

ovit

-0.13

 juvenile

-0.13

omite

-0.13

-stat

-0.13

POSITIVE LOGITS

cheid

0.17

Ħ

0.16

bens

0.16

istrovstvÃŃ

0.15

.sd

0.15

uali

0.14

tainment

0.14

Ã¨le

0.14

Å¾ÃŃ

0.14

ollect

0.14

Activations Density 0.005%

variations of the word "board" in different contexts

No Comments

No Known Activations