INDEX

Explanations

references to reports or issues regarding legal or forensic contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Lmfao

-0.72

tvguidetime

-0.63

 leveraging

-0.61

TLDR

-0.60

potentially

-0.56

izarro

-0.56

 transitioning

-0.54

ultuous

-0.54

});

-0.54

Wtf

-0.53

POSITIVE LOGITS

 yesterday

0.68

 Yesterday

0.61

ScopeManager

0.56

TintMode

0.55

Yesterday

0.55

 Programme

0.54

yesterday

0.54

 Meksiku

0.53

ən

0.53

Activations Density 0.124%

references to reports or issues regarding legal or forensic contexts

No Comments

No Known Activations