INDEX

Explanations

references to figures and graphics in a document

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

åŃ¦éĻ¢

-0.15

 Painter

-0.14

esar

-0.14

zones

-0.14

onto

-0.14

afe

-0.14

ighb

-0.14

elters

-0.14

cient

-0.13

 formal

-0.13

POSITIVE LOGITS

graphics

0.30

stand

0.20

box

0.16

stands

0.15

´Ī

0.15

iding

0.15

 Graphics

0.15

iders

0.15

´

0.14

 ruce

0.14

Activations Density 0.009%

references to figures and graphics in a document

No Comments

No Known Activations