INDEX

Explanations

the repetition of the word "to" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

forth

-0.13

Ð¸Ð½ÑĥÐ²

-0.13

 prefers

-0.13

iage

-0.13

DataContract

-0.13

ä¸ĬãģĴ

-0.13

afort

-0.12

alleng

-0.12

ulously

-0.12

POSITIVE LOGITS

 hear

0.33

 finally

0.31

 know

0.31

see

0.25

 behold

0.24

 have

0.24

finally

0.23

be

0.22

hear

0.22

 witness

0.22

Activations Density 0.066%

the repetition of the word "to" in various contexts

No Comments

No Known Activations