INDEX

Explanations

dialogue and interactions that involve questioning or discussions about knowledge and understanding

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

Speculation, doubt, or potential wrongdoing

np_acts-logits-general · gemini-2.0-flash

possibly faked

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

 seemingly

-0.40

 казалось

-0.40

NameValuePair

-0.35

agt

-0.35

 Bedürfn

-0.34

 aparentemente

-0.34

 apparently

-0.33

 Seem

-0.33

rood

-0.33

 bekannt

-0.33

POSITIVE LOGITS

 faked

0.66

possibly

0.65

Possibly

0.63

 faking

0.63

 possibly

0.63

 fake

0.63

 Possibly

0.63

 fraude

0.61

 belonged

0.60

 fakes

0.60

Activations Density 1.069%

dialogue and interactions that involve questioning or discussions about knowledge and understanding

Speculation, doubt, or potential wrongdoing

possibly faked

No Comments

No Known Activations