INDEX

Explanations

numerical identifiers and specifications related to products, likely ones used for technical purposes

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

vider

-0.16

ronics

-0.15

acam

-0.15

à¸¡à¸Ĥ

-0.14

ombat

-0.14

vÃ©

-0.14

ropol

-0.14

,SLOT

-0.13

 bÃło

-0.13

iani

-0.13

POSITIVE LOGITS

ipeg

0.14

pup

0.14

ison

0.14

 Teacher

0.14

iture

0.14

ennent

0.14

cli

0.14

arden

0.13

spy

0.13

 Torah

0.13

Activations Density 0.037%

numerical identifiers and specifications related to products, likely ones used for technical purposes

No Comments

No Known Activations