INDEX

Explanations

references to locations and places to visit or eat

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

agoon

-0.15

emez

-0.15

Sed

-0.15

itals

-0.15

wand

-0.14

edback

-0.14

ufe

-0.14

ìģ

-0.14

Ã¡genes

-0.14

stad

-0.14

POSITIVE LOGITS

ç½²

0.17

 escape

0.16

 retire

0.16

 Friedman

0.16

 holiday

0.16

 retir

0.16

 escaping

0.15

 getaway

0.15

 safely

0.15

ÑĨÐ¾

0.15

Activations Density 0.090%

references to locations and places to visit or eat

No Known Activations