INDEX

Explanations

references to collaborative contributions or edits in content discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

-0.16

enko

-0.16

ymous

-0.15

 fascism

-0.15

broken

-0.14

bers

-0.14

Zub

-0.14

ystery

-0.13

orial

-0.13

 Epstein

-0.13

POSITIVE LOGITS

ruz

0.18

è£ķ

0.17

arn

0.15

amax

0.14

ä¸Ī

0.14

ulado

0.14

eos

0.14

tele

0.13

ACKET

0.13

ÑĩÐ°Ð¹

0.13

Activations Density 0.004%

references to collaborative contributions or edits in content discussions

No Comments

No Known Activations