INDEX

Explanations

various forms of the word "of" in different contexts

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

 Lamp

-0.14

elle

-0.14

ied

-0.14

 Lucas

-0.14

ÛĮØ§

-0.14

ia

-0.13

chein

-0.13

iaux

-0.13

RS

-0.13

toPromise

-0.13

POSITIVE LOGITS

 sorts

0.18

FFE

0.15

LOPT

0.15

ecta

0.15

loff

0.15

ekt

0.14

witter

0.14

anka

0.14

ipi

0.14

Ł

0.14

Activations Density 0.064%

various forms of the word "of" in different contexts

No Comments

No Known Activations