INDEX

Explanations

mentions of the word "Aust" and its variations, which likely relate to Australia or Austrian topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

orgh

-0.20

jon

-0.15

urge

-0.15

lez

-0.15

dez

-0.15

ÑģÑĤÐ²ÐµÐ½

-0.15

ÑĢÐ¸Ð²Ð°

-0.15

shan

-0.15

izations

-0.14

bed

-0.14

POSITIVE LOGITS

enas

0.18

 Aust

0.17

chwitz

0.17

ensibly

0.17

ries

0.17

asia

0.17

umn

0.16

roph

0.16

èĮ

0.15

bery

0.15

Activations Density 0.006%

mentions of the word "Aust" and its variations, which likely relate to Australia or Austrian topics

No Comments

No Known Activations