INDEX

Explanations

repeated occurrences of the letter 'w' and variations of the word "wear" 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 @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

-0.17

-0.16

inÃ§

-0.15

ITHER

-0.15

íĺ

-0.14

ARGV

-0.14

 ÑĢÑı

-0.14

Haz

-0.14

æ¯Ĵ

-0.14

POSITIVE LOGITS

omens

0.20

iew

0.19

oment

0.19

rok

0.18

oman

0.18

ares

0.18

Ord

0.17

ondrous

0.17

hole

0.17

IEW

0.17

Activations Density 0.030%

repeated occurrences of the letter 'w' and variations of the word "wear" in various forms

No Comments

No Known Activations