INDEX

Explanations

instances of the word "Ir" and its variations, indicating a focus on the character or subject referred to as "Ir."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

illet

-0.16

emer

-0.16

dana

-0.16

et

-0.15

extr

-0.15

omic

-0.15

owo

-0.15

arda

-0.15

 evil

-0.14

obox

-0.14

POSITIVE LOGITS

regular

0.24

ving

0.21

replace

0.20

win

0.19

anian

0.19

ritable

0.18

landa

0.17

ises

0.17

irr

0.17

Activations Density 0.009%

instances of the word "Ir" and its variations, indicating a focus on the character or subject referred to as "Ir."

No Comments

No Known Activations

instances of the word "Ir" and its variations, indicating a focus on the character or subject referred to as "Ir."

No Comments

No Known Activations