INDEX

Explanations

numerical values or references to measurements and levels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 èĲ

-0.15

IMPLEMENT

-0.15

hart

-0.14

odied

-0.14

bate

-0.14

Ĥæķ°

-0.14

busters

-0.14

avou

-0.14

Úĺ

-0.14

ForRow

-0.14

POSITIVE LOGITS

yster

0.17

perature

0.15

uales

0.15

anko

0.14

oup

0.14

afil

0.14

 Watt

0.14

Ø§Ø¹Øª

0.14

sdl

0.14

 priv

0.14

Activations Density 0.000%

numerical values or references to measurements and levels

No Comments

No Known Activations