INDEX

Explanations

words that refer to specific letters in the alphabet, especially 'G' and variations of it

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

tk

-0.21

ÐµÐ½

-0.20

ett

-0.20

rid

-0.19

ui

-0.19

et

-0.17

ettle

-0.17

uh

-0.17

ame

-0.17

NU

-0.17

POSITIVE LOGITS

opher

0.22

rooms

0.18

rafted

0.18

localized

0.18

libc

0.18

nosis

0.18

 ener

0.18

KD

0.17

azing

0.17

urus

0.17

Activations Density 0.162%

words that refer to specific letters in the alphabet, especially 'G' and variations of it

No Comments

No Known Activations