INDEX

Explanations

terms related to specific entities or subjects of importance

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

aise

-0.15

SessionFactory

-0.15

adelphia

-0.14

imeo

-0.14

reative

-0.14

_Common

-0.14

ä»ģ

-0.14

 Matrix

-0.14

ief

-0.13

erman

-0.13

POSITIVE LOGITS

hoff

0.14

trx

0.14

rÃ¡

0.14

olta

0.14

amt

0.14

ichte

0.14

enza

0.14

benh

0.14

arsers

0.13

ITA

0.13

Activations Density 0.076%

terms related to specific entities or subjects of importance

No Comments

No Known Activations