INDEX

Explanations

actions related to gaining or losing control in specific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

oria

-0.15

dee

-0.15

sville

-0.15

abis

-0.15

 newArr

-0.14

bish

-0.14

ween

-0.14

 Mime

-0.14

iero

-0.14

 å¯

-0.13

POSITIVE LOGITS

ath

0.17

Ø§ÛĮØ´

0.16

ixel

0.16

Compra

0.15

aget

0.15

ĵĺ

0.15

atore

0.15

RT

0.14

ioc

0.14

bell

0.14

Activations Density 0.015%

actions related to gaining or losing control in specific contexts

No Comments

No Known Activations