INDEX

Explanations

instances of the word "went" or variations of it in descriptions of movement or transition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

achat

-0.20

forme

-0.16

echa

-0.16

isu

-0.16

ëĦ·

-0.15

 revolution

-0.15

Unsafe

-0.14

ãĥ¼ãĥĭ

-0.14

alion

-0.14

ewed

-0.14

POSITIVE LOGITS

on

0.19

 onto

0.19

to

0.19

worth

0.19

 from

0.16

 into

0.16

ihad

0.15

 proceeded

0.15

 Missing

0.14

 missing

0.14

Activations Density 0.042%

instances of the word "went" or variations of it in descriptions of movement or transition

No Comments

No Known Activations