INDEX

Explanations

versions and specifications of software or technology products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

nul

-0.17

Jal

-0.16

ometric

-0.16

 equiv

-0.15

ernes

-0.15

icker

-0.15

aland

-0.15

angs

-0.15

ÑĢÐ¸ÑģÑĤÐ¸

-0.14

ru

-0.14

POSITIVE LOGITS

rier

0.18

 Robbins

0.15

linger

0.15

gesi

0.15

ÏħÎ³

0.15

AGES

0.15

ä¸²

0.15

berger

0.15

leaf

0.14

auge

0.14

Activations Density 0.044%

versions and specifications of software or technology products

No Comments

No Known Activations