INDEX

Explanations

references to specific character names and attributes in animated series or films

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 prostituerade

-0.15

Î¼ÏĢ

-0.14

abh

-0.14

da

-0.14

omial

-0.13

 fetisch

-0.13

ousel

-0.13

 Craw

-0.13

oup

-0.13

ibr

-0.13

POSITIVE LOGITS

Az

0.30

Az

0.28

Meg

0.27

Meg

0.25

 Mind

0.24

Ez

0.22

Mag

0.22

meg

0.21

 Tart

0.21

az

0.20

Activations Density 0.002%

references to specific character names and attributes in animated series or films

No Comments

No Known Activations