INDEX

Explanations

references to specific locations or settings in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

apult

-0.15

 straw

-0.15

 Corner

-0.15

Zam

-0.14

Corner

-0.14

 duty

-0.14

 corner

-0.14

\Mapping

-0.14

 Straw

-0.14

readcr

-0.14

POSITIVE LOGITS

unh

0.15

Geh

0.15

pts

0.15

aylor

0.15

ENCY

0.15

agher

0.15

 Agility

0.14

ycastle

0.14

ge

0.14

enza

0.14

Activations Density 0.434%

references to specific locations or settings in various contexts

No Comments

No Known Activations