INDEX

Explanations

references to pieces of items or components described in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 ExecuteAsync

-0.85

esercito

-0.82

 Hald

-0.77

LookAnd

-0.76

 SIMBAD

-0.76

pośred

-0.74

IntentFilter

-0.74

Попис

-0.74

ullable

-0.73

(!__

-0.72

POSITIVE LOGITS

 pieces

2.59

 piece

2.58

 Pieces

2.43

 Piece

2.41

Pieces

2.35

 PIECE

2.31

Piece

2.28

pieces

2.19

piece

2.16

PIECE

1.75

Activations Density 0.048%

references to pieces of items or components described in various contexts

No Comments

No Known Activations