INDEX

Explanations

marks of punctuation and formatting symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 camer

-0.16

 editorial

-0.15

Uvs

-0.15

 artist

-0.15

Ã¢b

-0.14

 multimedia

-0.14

 pictures

-0.13

 video

-0.13

 portrait

-0.13

PostalCodes

-0.13

POSITIVE LOGITS

Directed

0.19

Production

0.19

Collider

0.19

Producer

0.18

actors

0.17

Characters

0.17

Movies

0.17

TV

0.17

DVD

0.17

 CAST

0.17

Activations Density 0.130%

marks of punctuation and formatting symbols

No Comments

No Known Activations