INDEX

Explanations

dialogue or conversational exchanges between speakers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 fencing

-0.17

oras

-0.16

ena

-0.14

Ãºng

-0.14

Ð°ÑĢÑħ

-0.14

aland

-0.14

 fences

-0.13

åĸ

-0.13

åİļ

-0.13

 openings

-0.13

POSITIVE LOGITS

Ã¼nd

0.16

eus

0.15

dden

0.15

udiantes

0.14

/TT

0.14

<?,

0.14

åĪĳ

0.14

('__

0.14

ĳ

0.14

Activations Density 0.013%

dialogue or conversational exchanges between speakers

No Comments

No Known Activations