INDEX

Explanations

references to the word "pearl" 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 @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ess

-0.16

ew

-0.16

esc

-0.15

ya

-0.15

lar

-0.15

ap

-0.15

alf

-0.15

iji

-0.14

avec

-0.14

 bald

-0.14

POSITIVE LOGITS

izoph

0.16

toi

0.16

ë°ľ

0.15

èĢĢ

0.14

asso

0.14

 sÃ»

0.14

uzzer

0.14

 nicotine

0.14

 Hayward

0.14

axon

0.14

Activations Density 0.006%

references to the word "pearl" and its variations

No Comments

No Known Activations