INDEX

Explanations

instances of the word "In" and variations thereof

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ellation

-0.15

dll

-0.15

tras

-0.15

uhe

-0.14

dictions

-0.14

eling

-0.14

dre

-0.14

rvÃ©

-0.13

 hoch

-0.13

dney

-0.13

POSITIVE LOGITS

ÑĤÐ°Ð¿

0.15

INO

0.15

RowAt

0.14

oven

0.14

Ð»Ð°Ð´Ð¸

0.14

åĬĥ

0.14

 receipt

0.14

Additionally

0.14

 razor

0.14

åĪĴ

0.14

Activations Density 0.066%

instances of the word "In" and variations thereof

No Comments

No Known Activations