INDEX

Explanations

mentions of roads and road-related terms

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

 незавершена

-0.62

FailureListener

-0.58

cault

-0.56

 Peers

-0.56

 bira

-0.55

Népesség

-0.54

 Denison

-0.53

 Utile

-0.52

lesque

-0.52

Composable

-0.51

POSITIVE LOGITS

 Road

4.65

Road

4.26

 road

4.11

 ROAD

3.97

road

3.84

ROAD

3.50

Rd

3.38

 Roads

2.89

 roads

2.84

Roads

2.50

Activations Density 0.062%

mentions of roads and road-related terms

No Comments

No Known Activations