INDEX

Explanations

references to completeness or totality in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

emia

-0.18

ennes

-0.18

ette

-0.17

ess

-0.16

off

-0.16

zelf

-0.15

es

-0.15

ema

-0.15

jem

-0.15

estre

-0.15

POSITIVE LOGITS

/full

0.23

-ÑĤÐ°ÐºÐ¸

0.18

IRCLE

0.18

DonaldTrump

0.17

filled

0.17

tsky

0.17

erton

0.16

lege

0.16

-bodied

0.16

sville

0.15

Activations Density 0.040%

references to completeness or totality in various contexts

No Comments

No Known Activations