INDEX

Explanations

instances of the word "can" in various forms

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

enburg

-0.18

-0.16

oker

-0.15

mint

-0.15

osci

-0.15

gam

-0.15

duk

-0.14

gem

-0.14

 Phase

-0.14

 vera

-0.14

POSITIVE LOGITS

!=(

0.18

Ð°ÑĢÑħ

0.16

à¸¶à¸ģ

0.15

 yetiÅŁtir

0.15

ìĸ

0.14

chine

0.14

vrier

0.14

ederland

0.14

=start

0.14

chie

0.14

Activations Density 0.046%

instances of the word "can" in various forms

No Comments

No Known Activations