INDEX

Explanations

variations of the word "American."

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

infeld

-0.15

IRS

-0.15

éĻĲ

-0.15

ivy

-0.15

Lol

-0.15

 Lars

-0.14

ython

-0.14

/goto

-0.14

chief

-0.13

ewise

-0.13

POSITIVE LOGITS

icana

0.19

sterdam

0.19

ikan

0.18

ican

0.18

ica

0.18

ika

0.18

ijken

0.17

igan

0.16

ICA

0.16

rine

0.16

Activations Density 0.011%

variations of the word "American."

No Comments

No Known Activations