INDEX

Explanations

instances of the word "interact."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ugas

-0.19

ald

-0.16

acter

-0.15

iske

-0.15

hand

-0.15

aret

-0.14

Dro

-0.14

 Bender

-0.14

 pron

-0.14

 tháº£o

-0.13

POSITIVE LOGITS

-prepend

0.18

edd

0.17

Ä±rak

0.16

otron

0.15

dum

0.15

antan

0.14

locker

0.14

ologie

0.14

inati

0.13

otland

0.13

Activations Density 0.006%

instances of the word "interact."

No Comments

No Known Activations