INDEX

Explanations

statements and quotations expressed by individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Angeles

-0.73

CUIT

-0.68

geführt

-0.68

 Wilfred

-0.67

την

-0.67

Manfred

-0.66

 Selamat

-0.64

 Ortiz

-0.64

 franca

-0.64

bled

-0.63

POSITIVE LOGITS

Says

1.04

SAY

0.98

 Says

0.98

er

0.97

Cay

0.97

Nye

0.97

 says

0.97

Nye

0.92

 HasFactory

0.89

Say

0.88

Activations Density 0.079%

statements and quotations expressed by individuals

No Comments

No Known Activations