INDEX

Explanations

adjectives and adverbs indicating quality or effectiveness

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

increments

-0.17

Slinky

-0.16

 {{--<

-0.15

ivec

-0.15

 erotique

-0.14

 addCriterion

-0.14

.FC

-0.14

orang

-0.14

amac

-0.14

.fc

-0.14

POSITIVE LOGITS

ast

0.24

has

0.22

has

0.21

ad

0.20

 quanto

0.19

 than

0.19

0.18

 anybody

0.18

ae

0.17

HAS

0.17

Activations Density 0.044%

adjectives and adverbs indicating quality or effectiveness

No Comments

No Known Activations