INDEX

Explanations

actions involving mixing or applying substances, particularly in cooking or cleaning contexts

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

.usermodel

-0.15

 vÃŃde

-0.15

fra

-0.14

 thanh

-0.14

arel

-0.14

amba

-0.14

 ÙħÙĪØ¬

-0.14

çĤİ

-0.14

grily

-0.14

heim

-0.14

POSITIVE LOGITS

 baking

0.24

tal

0.24

cit

0.23

WD

0.23

 rubbing

0.23

 petroleum

0.22

 ment

0.22

 Formal

0.22

 formal

0.21

gly

0.21

Activations Density 0.323%

actions involving mixing or applying substances, particularly in cooking or cleaning contexts

No Comments

No Known Activations