INDEX

Explanations

negative prefixes that signify a lack or absence of something

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

Ð¾Ð¿Ð°Ñģ

-0.16

à¸§à¸ĩ

-0.14

ç§

-0.14

 [âĢ¦]↵↵

-0.14

roker

-0.14

moduleId

-0.14

Ð¾Ð±ÑĢÐ°Ð·

-0.14

'gc

-0.13

unker

-0.13

/***/

-0.13

POSITIVE LOGITS

unate

0.16

cov

0.16

ogan

0.15

éĩı

0.15

 bottom

0.15

oji

0.14

lingen

0.14

 tack

0.14

ancellable

0.14

LTR

0.14

Activations Density 0.152%

negative prefixes that signify a lack or absence of something

No Comments

No Known Activations