INDEX

Explanations

references to HTML documents and coding elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

dl

-0.17

oni

-0.15

ett

-0.14

avad

-0.14

olas

-0.14

Ã¸d

-0.14

ess

-0.14

ette

-0.13

Commerce

-0.13

ety

-0.13

POSITIVE LOGITS

ilde

0.16

ODEV

0.16

lesc

0.15

cene

0.15

sandbox

0.14

scape

0.14

endoza

0.13

DivElement

0.13

finity

0.13

Activations Density 0.033%

references to HTML documents and coding elements

No Comments

No Known Activations