INDEX

Explanations

various forms of punctuation and dashes indicative of breaks or changes in thought

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

acman

-0.16

eed

-0.15

SOLE

-0.14

buquerque

-0.14

ause

-0.14

hl

-0.14

ighton

-0.14

orque

-0.14

ellido

-0.13

adolu

-0.13

POSITIVE LOGITS

oret

0.17

-uppercase

0.16

ãĤĵãģ¨

0.15

 pornografia

0.15

0.13

 both

0.13

.TestTools

0.13

ayah

0.13

Tat

0.13

 ones

0.13

Activations Density 0.130%

various forms of punctuation and dashes indicative of breaks or changes in thought

No Comments

No Known Activations