INDEX

Explanations

references to personal or possessive pronouns in relation to particular subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ipar

-0.15

.emit

-0.15

nelle

-0.15

ç°

-0.14

marshall

-0.14

á»ĳc

-0.14

Invariant

-0.13

SES

-0.13

æ³¢

-0.13

anas

-0.13

POSITIVE LOGITS

 finances

0.16

getDisplay

0.15

 behalf

0.15

acus

0.15

enou

0.14

 Criterion

0.14

 possible

0.13

 ÄĳÃŃch

0.13

 inter

0.13

 possibile

0.13

Activations Density 0.174%

references to personal or possessive pronouns in relation to particular subjects

No Comments

No Known Activations