INDEX

Explanations

phrases indicating significant attributes or effects within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_13/width_16k/average_l0_132

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

 fleste

-0.45

 themselves

-0.43

their

-0.43

 their

-0.43

themselves

-0.42

Their

-0.41

 mají

-0.41

 mają

-0.41

 they

-0.39

 Their

-0.39

POSITIVE LOGITS

 itself

0.74

Autoritní

0.72

 himself

0.69

itself

0.65

himself

0.61

 himſelf

0.54

таратура

0.53

 Itself

0.53

 Himself

0.53

Activations Density 1.294%

phrases indicating significant attributes or effects within a context

No Comments

No Known Activations