INDEX

Explanations

references to policies and guidelines

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mystérie

-0.44

withCredentials

-0.44

 aveug

-0.40

saraba

-0.40

Find

-0.38

 kapturem

-0.38

 אית

-0.38

 dedans

-0.38

Egyptian

-0.37

The

-0.36

POSITIVE LOGITS

 POLICY

1.16

 policy

1.10

 Policy

1.09

policy

1.07

Policy

1.02

POLICY

1.02

 Policies

0.89

 POLICIES

0.89

 policies

0.87

Policies

0.84

Activations Density 0.149%

references to policies and guidelines

No Comments

No Known Activations