INDEX

Explanations

specific articles and common nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 itſelf

-1.25

 myſelf

-1.13

 purpoſe

-1.06

 themſelves

-1.06

 raiſ

-1.04

 ſtate

-1.03

 fubject

-1.00

 himſelf

-0.99

 preſent

-0.96

Efq

-0.96

POSITIVE LOGITS

the

1.23

The

1.13

The

0.96

the

0.90

die

0.85

der

0.84

0.78

den

0.76

Οι

0.75

 οι

0.73

Activations Density 0.043%

specific articles and common nouns

No Comments

No Known Activations