INDEX

Explanations

information about privacy and marketing preferences

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

-0.16

omu

-0.15

ve

-0.15

ple

-0.15

ozem

-0.14

munition

-0.14

ocalypse

-0.14

ourke

-0.14

lea

-0.14

kip

-0.14

POSITIVE LOGITS

afi

0.15

ercial

0.15

Os

0.15

Î»Î¼

0.15

igli

0.14

ToObject

0.14

AppName

0.14

 Fathers

0.13

ÄĽÅĻ

0.13

avaÅŁ

0.13

Activations Density 0.018%

information about privacy and marketing preferences

No Comments

No Known Activations