INDEX

Explanations

references to specific products and their characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

-0.17

etur

-0.17

"crypto

-0.16

ocial

-0.14

ãĤ¦ãĥ³

-0.14

ãĤ¤ãĥ¤

-0.14

mart

-0.13

uw

-0.13

kar

-0.13

POSITIVE LOGITS

iquer

0.15

Î¼Îµ

0.15

criptor

0.14

PIP

0.14

 Cannon

0.14

itivity

0.14

Paid

0.14

Ø³Ùĥ

0.14

itized

0.14

Activations Density 0.115%

references to specific products and their characteristics

No Comments

No Known Activations