INDEX

Explanations

mentions of the concept of occurrence or events happening

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

coming

-0.18

 cÃ²n

-0.16

igers

-0.16

elson

-0.15

held

-0.15

oria

-0.15

castle

-0.15

tery

-0.15

ym

-0.14

arna

-0.14

POSITIVE LOGITS

ances

0.21

ally

0.19

ance

0.16

ration

0.16

uate

0.16

stance

0.15

urs

0.15

adal

0.15

ional

0.15

uating

0.14

Activations Density 0.025%

mentions of the concept of occurrence or events happening

No Comments

No Known Activations