INDEX

Explanations

references to windows and openings in physical spaces

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

athom

-0.18

á»Ľp

-0.16

ł

-0.15

ALIGN

-0.15

hurst

-0.15

ownik

-0.15

rypt

-0.15

.Alignment

-0.14

 sÃ¸ger

-0.14

âĶĶ

-0.14

POSITIVE LOGITS

/window

0.16

 windows

0.16

affen

0.15

rones

0.14

ìŀ¬

0.14

fen

0.14

uft

0.14

ess

0.14

 Mund

0.14

rias

0.14

Activations Density 0.090%

references to windows and openings in physical spaces

No Comments

No Known Activations