INDEX

Explanations

terms indicating previous names or identities of people, places, or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

insky

-0.15

Ø¨Ø§Ø¯

-0.14

UILTIN

-0.14

.epam

-0.14

->[

-0.13

iculty

-0.13

yy

-0.13

_Callback

-0.13

ptrdiff

-0.13

una

-0.12

POSITIVE LOGITS

 Former

0.17

plib

0.16

pond

0.15

formerly

0.14

 expend

0.14

į¼

0.14

áº©n

0.14

 repmat

0.14

cket

0.14

Activations Density 0.016%

terms indicating previous names or identities of people, places, or entities

No Comments

No Known Activations