INDEX

Explanations

numerical data or figures, likely related to measurements or quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

enger

-0.16

ubb

-0.16

 Doyle

-0.14

oble

-0.14

nda

-0.14

enden

-0.13

ouched

-0.13

oram

-0.13

 readonly

-0.13

allon

-0.13

POSITIVE LOGITS

é¼

0.15

Ø³ÙĬØ©

0.14

ieber

0.14

 Coupe

0.14

sav

0.14

Sax

0.14

 Home

0.14

Ãºsqueda

0.13

prung

0.13

/Home

0.13

Activations Density 0.110%

numerical data or figures, likely related to measurements or quantities

No Comments

No Known Activations