INDEX

Explanations

forms of the word "one" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

agens

-0.19

itele

-0.17

.mk

-0.15

ä»¤

-0.15

ernes

-0.15

ificates

-0.15

ovna

-0.15

utzer

-0.14

æĶ

-0.14

dat

-0.14

POSITIVE LOGITS

.wr

0.15

alist

0.14

á»ģn

0.14

Knife

0.14

 Britt

0.14

 Freund

0.14

isper

0.14

seealso

0.13

ableObject

0.13

wit

0.13

Activations Density 0.007%

forms of the word "one" in various contexts

No Comments

No Known Activations