INDEX

Explanations

instances of the word "when."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

them

-0.19

ise

-0.17

orsi

-0.15

 tieten

-0.15

unately

-0.15

 herself

-0.15

ize

-0.14

 Ð¿Ð¾ÐºÐ°

-0.14

ly

-0.14

ally

-0.14

POSITIVE LOGITS

/if

0.47

soever

0.43

EVER

0.33

 they

0.32

 faced

0.31

 asked

0.29

it

0.29

we

0.28

/how

0.28

 compared

0.27

Activations Density 0.138%

instances of the word "when."

No Comments

No Known Activations