INDEX

Explanations

instances of the word "not" and itsNegations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

utenberg

-0.15

ä¹Ī

-0.14

papers

-0.14

åĺī

-0.14

hed

-0.14

Å¡t

-0.14

 Gilbert

-0.13

letters

-0.13

ei

-0.13

POSITIVE LOGITS

obe

0.19

 è¢

0.15

 alike

0.15

ucch

0.15

ocos

0.14

teri

0.14

å¾®è½¯éĽħé»ĳ

0.14

atform

0.14

redo

0.14

UPER

0.14

Activations Density 0.038%

instances of the word "not" and itsNegations

No Comments

No Known Activations