INDEX

Explanations

references to individuals, specifically their names and pronouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

į

-0.17

Äįin

-0.16

aise

-0.14

 cuer

-0.14

ameron

-0.14

rawl

-0.14

htag

-0.14

icum

-0.14

 dosage

-0.14

ÑĢÐ°Ð½Ð°

-0.14

POSITIVE LOGITS

kses

0.15

itou

0.15

obec

0.15

ÄĽt

0.14

ÐĺÐ¢

0.14

vos

0.14

 KÃ¼r

0.13

Ú©Ø²

0.13

addtogroup

0.13

bedo

0.13

Activations Density 0.274%

references to individuals, specifically their names and pronouns

No Comments

No Known Activations