INDEX

Explanations

references to the word "bat" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

hoff

-0.21

orgh

-0.17

idge

-0.16

parator

-0.16

Vit

-0.16

adoo

-0.16

oras

-0.15

.nih

-0.15

iel

-0.15

edList

-0.15

POSITIVE LOGITS

htub

0.27

avia

0.24

Bat

0.23

ting

0.23

wing

0.22

Bat

0.21

bat

0.20

tement

0.19

leground

0.19

ista

0.19

Activations Density 0.012%

references to the word "bat" and its variations

No Comments

No Known Activations