INDEX

Explanations

sections or headings 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 @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

enci

-0.18

cast

-0.15

ÙĨØ©

-0.14

ignKey

-0.14

obo

-0.14

SCI

-0.14

ssi

-0.13

uild

-0.13

uesta

-0.13

POSITIVE LOGITS

Locker

0.15

igr

0.14

",-

0.14

 Weston

0.14

LOCK

0.14

 Suites

0.14

Gef

0.14

Åĳ

0.14

 Malone

0.14

 pursued

0.13

Activations Density 0.005%

sections or headings in a document

No Comments

No Known Activations