INDEX

Explanations

words and phrases associated with taste and flavor

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

eer

-0.17

-0.16

eck

-0.15

 wedge

-0.15

elem

-0.15

-0.14

CRET

-0.14

irk

-0.14

 Escort

-0.14

ared

-0.14

POSITIVE LOGITS

ework

0.36

eness

0.34

eland

0.32

eman

0.31

ename

0.31

edef

0.31

ereg

0.31

ewith

0.30

eward

0.30

eway

0.30

Activations Density 0.256%

words and phrases associated with taste and flavor

No Comments

No Known Activations