INDEX

Explanations

proper names and brand names related to products, specifically in the context of media and crafting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ynos

-0.17

è©

-0.15

thon

-0.15

à¸¢à¸²à¸Ļ

-0.15

hausen

-0.14

cÃ¼

-0.14

onium

-0.14

 autof

-0.14

itere

-0.14

opard

-0.14

POSITIVE LOGITS

owy

0.17

bare

0.17

 bare

0.16

 myself

0.16

olu

0.15

0.14

 Bare

0.14

 Hector

0.14

gro

0.14

 Summers

0.14

Activations Density 0.034%

proper names and brand names related to products, specifically in the context of media and crafting

No Comments

No Known Activations