INDEX

Explanations

references to floors and buildings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

opia

-0.15

endar

-0.15

 Ø§ÙĦØ¬Ùħ

-0.14

byn

-0.14

ooth

-0.14

à¹Īà¸Ńà¸ĩ

-0.14

Ã©cial

-0.14

kop

-0.14

ë§¥

-0.13

ëł¥

-0.13

POSITIVE LOGITS

 floor

0.84

 floors

0.79

 Floor

0.76

floor

0.71

-floor

0.69

Floor

0.67

 Floors

0.64

_floor

0.56

.floor

0.50

 flooring

0.49

Activations Density 0.136%

references to floors and buildings

No Comments

No Known Activations