INDEX

Explanations

terms related to utility and universal concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ald

-0.19

edly

-0.16

ed

-0.16

sb

-0.15

lett

-0.15

iw

-0.15

ening

-0.15

ish

-0.15

ingo

-0.15

stral

-0.15

POSITIVE LOGITS

ized

0.21

izing

0.19

ization

0.19

 dá»¥ng

0.17

ally

0.16

alse

0.15

ALLY

0.15

idades

0.15

phem

0.15

ocity

0.14

Activations Density 0.053%

terms related to utility and universal concepts

No Comments

No Known Activations