INDEX

Explanations

references to firearms and the term "gun"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

oha

-0.17

ÐµÑĢÐ¸Ð½

-0.16

ÑĢÐ°Ð¶Ð´

-0.15

alling

-0.15

urnal

-0.15

©

-0.15

cue

-0.14

usz

-0.14

ustin

-0.14

ally

-0.14

POSITIVE LOGITS

pow

0.28

ny

0.21

ning

0.21

ned

0.20

linger

0.20

shots

0.19

erals

0.19

metal

0.18

nen

0.18

ners

0.17

Activations Density 0.024%

references to firearms and the term "gun"

No Comments

No Known Activations