INDEX

Explanations

references to formal titles and specific dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

 compass

-0.17

 subs

-0.16

ottes

-0.15

-0.14

 facts

-0.14

 banners

-0.14

EXIT

-0.14

 rank

-0.14

 subt

-0.14

POSITIVE LOGITS

.secret

0.15

à¸¤à¸©

0.15

 SECRET

0.15

 Sink

0.15

 ÑģÐµÐºÑĢÐµÑĤ

0.14

rored

0.14

adin

0.14

oreach

0.14

 eskort

0.14

 ÑĤÐ¾Ð²

0.14

Activations Density 0.035%

references to formal titles and specific dates

No Comments

No Known Activations