INDEX

Explanations

references to specific names or entities within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

anders

-0.21

ohen

-0.16

LineColor

-0.15

igkeit

-0.15

.Logger

-0.14

 bent

-0.14

ave

-0.14

utsch

-0.14

Ã¼st

-0.14

.Toolkit

-0.13

POSITIVE LOGITS

aldo

0.18

iste

0.17

rious

0.17

ionage

0.17

chie

0.17

ka

0.16

IOD

0.16

alog

0.16

pio

0.15

ions

0.15

Activations Density 0.018%

references to specific names or entities within a context

No Comments

No Known Activations