INDEX

Explanations

instances of ongoing actions or activities

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

ena

-0.15

ignum

-0.15

Ce

-0.15

åĲĪåĲĮ

-0.15

Adv

-0.15

ÑĢÐ¾ÑĤ

-0.14

 damage

-0.14

awa

-0.14

_dirty

-0.14

POSITIVE LOGITS

ãģĭãģĳ

0.18

0.16

vester

0.15

uilder

0.15

enido

0.14

agento

0.14

 Gotham

0.14

 ÑĩÐ¸Ð½

0.14

pul

0.14

VEST

0.14

Activations Density 0.212%

instances of ongoing actions or activities

No Comments

No Known Activations