INDEX

Explanations

occurrences of female character titles or names, particularly those starting with "Mrs."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

NSK

-0.14

cio

-0.14

adiens

-0.14

girl

-0.14

mada

-0.14

ê¶Į

-0.14

-girl

-0.14

 GURL

-0.13

ersonic

-0.13

 tranh

-0.13

POSITIVE LOGITS

 Grund

0.17

iggins

0.17

 America

0.17

America

0.17

enor

0.16

liable

0.15

enschaft

0.15

seau

0.15

 Claus

0.15

 Doub

0.15

Activations Density 0.037%

occurrences of female character titles or names, particularly those starting with "Mrs."

No Comments

No Known Activations