INDEX

Explanations

references to software and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

oku

-0.18

atty

-0.16

gom

-0.16

ehr

-0.15

att

-0.15

gon

-0.14

Ãªte

-0.14

nw

-0.14

reak

-0.14

 Throw

-0.13

POSITIVE LOGITS

 distributed

0.14

IRC

0.14

orts

0.14

 Andrews

0.14

 publicly

0.13

 Reeves

0.13

itoris

0.13

 Ð²Ð¾Ð¹Ð½Ñĭ

0.13

 spol

0.13

brace

0.13

Activations Density 0.003%

references to software and related terms

No Comments

No Known Activations