INDEX

Explanations

instances of the words "have" and "has" in various forms and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Buckley

-0.16

IIIK

-0.15

erde

-0.14

Ð¾ÑĢÐ¼

-0.14

saw

-0.14

ÑģÑĸÐ¼

-0.14

enburg

-0.14

 knew

-0.14

wich

-0.14

 yapÄ±lÄ±r

-0.14

POSITIVE LOGITS

 been

0.61

been

0.51

 Been

0.40

 BEEN

0.38

be

0.38

ben

0.37

bee

0.36

Been

0.36

 Ã©tÃ©

0.35

 sido

0.34

Activations Density 0.146%

instances of the words "have" and "has" in various forms and contexts

No Comments

No Known Activations