INDEX

Explanations

navigation elements and links in a web document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ynos

-0.18

ÑħÑĥ

-0.18

BorderStyle

-0.16

deaux

-0.16

ocab

-0.15

aternity

-0.15

mium

-0.15

BALL

-0.15

ahoo

-0.15

ainer

-0.14

POSITIVE LOGITS

aris

0.15

arium

0.15

far

0.14

_PIXEL

0.13

 Pink

0.13

autorelease

0.13

lis

0.13

 personal

0.13

 Literature

0.13

bx

0.13

Activations Density 0.015%

navigation elements and links in a web document

No Comments

No Known Activations