INDEX

Explanations

numerical values and measurements related to positions and dimensions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

UnderTest

-0.17

linger

-0.17

celed

-0.15

plode

-0.14

 typings

-0.14

ostel

-0.14

wh

-0.14

Å¼y

-0.14

vise

-0.13

fik

-0.13

POSITIVE LOGITS

unta

0.15

ewis

0.15

umont

0.14

 Hick

0.14

erring

0.14

ginas

0.14

etz

0.13

Inf

0.13

 eens

0.13

Activations Density 0.111%

numerical values and measurements related to positions and dimensions

No Comments

No Known Activations

numerical values and measurements related to positions and dimensions

No Comments

No Known Activations