INDEX

Explanations

words that relate to entertainment or media

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

zon

-0.19

uld

-0.17

ild

-0.16

iled

-0.15

ake

-0.15

ideon

-0.15

soever

-0.14

CurrentValue

-0.14

 Herc

-0.14

POSITIVE LOGITS

iaux

0.16

æ°ĳ

0.16

 Lives

0.16

urement

0.15

Å¼y

0.15

sgi

0.14

urb

0.14

 passe

0.14

croll

0.14

Dur

0.14

Activations Density 0.000%

words that relate to entertainment or media

No Comments

No Known Activations