INDEX

Explanations

terms related to testing and evaluation processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

isse

-0.17

ob

-0.16

isser

-0.16

ilda

-0.15

orno

-0.15

nad

-0.15

icket

-0.15

çĶ

-0.15

 Shape

-0.14

aiser

-0.14

POSITIVE LOGITS

gres

0.17

emos

0.16

abee

0.16

kaÃ§

0.15

ylene

0.15

ÃŃda

0.15

tec

0.15

heim

0.15

upert

0.14

appa

0.14

Activations Density 0.038%

terms related to testing and evaluation processes

No Comments

No Known Activations