INDEX

Explanations

phrases that describe conditions or qualities in a comparative manner

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

.habbo

-0.18

yonel

-0.17

nej

-0.15

urat

-0.15

zew

-0.15

iska

-0.15

eka

-0.14

Ð¾ÑĢÐ¾Ð´

-0.14

zte

-0.14

ICODE

-0.13

POSITIVE LOGITS

 that

0.23

that

0.18

 bahwa

0.18

Łèĥ½

0.17

 hogy

0.16

	that

0.16

©

0.15

Ãºa

0.15

rend

0.15

 dass

0.14

Activations Density 0.055%

phrases that describe conditions or qualities in a comparative manner

No Comments

No Known Activations