INDEX

Explanations

equal signs, suggesting it is identifying variable assignments or equations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

geber

-0.16

ipes

-0.15

 Ð·ÐµÐ¼Ð»Ñı

-0.14

Ð¾ÑĢÐ°Ð·

-0.14

HU

-0.14

assed

-0.14

vern

-0.13

rezent

-0.13

 breat

-0.13

POSITIVE LOGITS

itori

0.15

ulace

0.15

dition

0.14

apel

0.14

Caps

0.14

ipple

0.14

achs

0.14

bish

0.14

hek

0.14

Î¾Î·

0.13

Activations Density 0.020%

equal signs, suggesting it is identifying variable assignments or equations

No Comments

No Known Activations