INDEX

Explanations

titles or headings, particularly those that include the word "The."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ÃŃna

-0.17

aud

-0.16

nam

-0.15

sted

-0.15

okens

-0.14

Ã¢n

-0.13

Dispatch

-0.13

tokens

-0.13

kok

-0.13

hon

-0.13

POSITIVE LOGITS

 Vera

0.15

_AG

0.14

.localization

0.14

ombat

0.13

ACHE

0.13

 cuales

0.13

AZ

0.13

 dissoci

0.13

ACE

0.13

ace

0.13

Activations Density 0.100%

titles or headings, particularly those that include the word "The."

No Comments

No Known Activations