INDEX

Explanations

references to specific types of 'ac' related items or terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

zeug

-0.18

 itemprop

-0.15

erk

-0.15

auer

-0.15

retch

-0.14

orce

-0.14

ono

-0.14

inks

-0.14

esc

-0.14

POSITIVE LOGITS

ording

0.21

Ac

0.21

adian

0.21

/ac

0.20

ac

0.19

#ac

0.19

acias

0.19

acia

0.18

CELER

0.18

tual

0.17

Activations Density 0.017%

references to specific types of 'ac' related items or terms

No Comments

No Known Activations