INDEX

Explanations

references to incidents involving legal accusations and courtroom proceedings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

URNS

-0.15

 Nickel

-0.15

lip

-0.14

ntax

-0.14

 dispensaries

-0.14

 LANG

-0.14

lak

-0.14

prites

-0.14

girls

-0.13

Ð°ÑĤÐºÑĥ

-0.13

POSITIVE LOGITS

 Space

0.31

 Kevin

0.29

Space

0.27

 SPACE

0.25

Kevin

0.25

SPACE

0.23

 space

0.23

-space

0.21

uran

0.21

 House

0.21

Activations Density 0.001%

references to incidents involving legal accusations and courtroom proceedings

No Comments

No Known Activations