INDEX

Explanations

phrases or descriptors indicating quality or type that imply comparison or classification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

owell

-0.17

 bump

-0.16

LOB

-0.16

ieri

-0.15

 choice

-0.14

Alb

-0.14

oden

-0.14

Virgin

-0.14

Î³ÏĮ

-0.14

POSITIVE LOGITS

TextChanged

0.15

 Zucker

0.14

oric

0.14

iol

0.14

ingu

0.14

cha

0.14

ermal

0.14

eca

0.14

/INFO

0.14

hee

0.14

Activations Density 0.056%

phrases or descriptors indicating quality or type that imply comparison or classification

No Comments

No Known Activations