INDEX

Explanations

information related to pages and pagination

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

Pages

-0.19

self

-0.17

ry

-0.17

ship

-0.16

sole

-0.16

ively

-0.15

ly

-0.15

Page

-0.15

_pages

-0.15

ÑĢÐµÑģ

-0.15

POSITIVE LOGITS

antry

0.30

ant

0.24

views

0.21

/post

0.19

 ÄĳÃŃch

0.19

-load

0.18

peed

0.17

gy

0.17

aud

0.16

-turn

0.16

Activations Density 0.038%

information related to pages and pagination

No Comments

No Known Activations