INDEX

Explanations

positive adjectives that describe quality or performance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

æ£ĭçīĮ

-0.16

-0.15

affer

-0.15

cki

-0.15

ãĥ³ãĥĨãĤ£

-0.14

abbr

-0.14

SEQUENTIAL

-0.14

anie

-0.14

spender

-0.14

POSITIVE LOGITS

 impression

0.22

 impressions

0.21

job

0.21

avors

0.20

 favors

0.20

fav

0.19

 favor

0.18

favor

0.18

 disappearing

0.18

fav

0.18

Activations Density 0.031%

positive adjectives that describe quality or performance

No Comments

No Known Activations