INDEX

Explanations

phrases that highlight membership or identity within various roles or groups

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

iene

-0.17

ryn

-0.15

awn

-0.15

ople

-0.15

Ð»ÐµÐ¹

-0.14

anki

-0.14

aal

-0.14

avirus

-0.14

oten

-0.14

adia

-0.14

POSITIVE LOGITS

0.15

ergus

0.15

á»ģ

0.14

><?

0.14

uzzi

0.14

 Maher

0.14

 Morm

0.14

Ð¾Ð¼Ð°

0.14

 Ð¢ÑĥÑĤ

0.13

Activations Density 0.054%

phrases that highlight membership or identity within various roles or groups

No Comments

No Known Activations