INDEX

Explanations

references to gender-related biases and expectations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

opup

-0.19

osh

-0.15

urge

-0.15

ì§¸

-0.14

udu

-0.14

ator

-0.14

ocese

-0.14

angl

-0.14

ogan

-0.13

ÑĢÐ°Ð

-0.13

POSITIVE LOGITS

peg

0.15

enia

0.15

Interrupt

0.15

 Pruitt

0.14

oke

0.14

iyon

0.14

Ã¡tek

0.14

quia

0.13

§

0.13

Activations Density 0.091%

references to gender-related biases and expectations

No Comments

No Known Activations