INDEX

Explanations

specific proper nouns or names, particularly those related to individuals in entertainment or academia

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ijke

-0.20

äºŃ

-0.15

ongan

-0.15

seau

-0.15

polation

-0.15

erse

-0.15

.Chain

-0.14

xima

-0.14

opus

-0.14

evin

-0.14

POSITIVE LOGITS

atu

0.17

 Woodward

0.17

0.16

inge

0.16

 ineff

0.15

Alv

0.15

iu

0.15

0.14

 Woods

0.14

Activations Density 0.004%

specific proper nouns or names, particularly those related to individuals in entertainment or academia

No Comments

No Known Activations