INDEX

Explanations

references to various types of assets

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

 téléphonique

-0.71

 compréhen

-0.70

errHandler

-0.67

 dénon

-0.66

StandardCharsets

-0.65

 FetchType

-0.63

AndEndTag

-0.63

ussis

-0.62

 mukana

-0.61

expandindo

-0.60

POSITIVE LOGITS

 asset

3.84

 assets

3.56

 Asset

3.38

 Assets

3.36

asset

3.22

Asset

3.12

 ASSETS

2.83

Assets

2.79

assets

2.57

ASSET

2.46

Activations Density 0.080%

references to various types of assets

No Comments

No Known Activations