INDEX

Explanations

the presence of specific names or terms, particularly those that start with 'Ek' or related variants

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

tuk

-0.16

ìļ±

-0.14

ovol

-0.14

illac

-0.14

evity

-0.14

ÑĢÐ°Ð¹

-0.14

ighth

-0.14

unct

-0.14

agogue

-0.13

Forgery

-0.13

POSITIVE LOGITS

ker

0.26

kers

0.23

inesis

0.20

ki

0.19

ater

0.18

onom

0.18

ken

0.18

ky

0.18

tron

0.18

ran

0.18

Activations Density 0.024%

the presence of specific names or terms, particularly those that start with 'Ek' or related variants

No Comments

No Known Activations