INDEX

Explanations

numerical values, particularly in a data or coding context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

zych

-0.17

 Chung

-0.15

ayah

-0.15

Î³ÎºÎ±

-0.15

baugh

-0.15

eyer

-0.14

lbrace

-0.14

 Cout

-0.13

SplitOptions

-0.13

 Chase

-0.13

POSITIVE LOGITS

zap

0.18

asal

0.15

åĽº

0.15

ok

0.14

nak

0.14

append

0.14

atest

0.14

bak

0.14

PCA

0.14

Activations Density 0.056%

numerical values, particularly in a data or coding context

No Comments

No Known Activations