INDEX

Explanations

occurrences of specific formatting commands or notational symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 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

 HasFactory

-0.57

 nahilalakip

-0.56

脚注の使い方

-0.53

 zoude

-0.51

 wikipagina

-0.51

 <<<<<<<<<<<<<<

-0.50

OCCURRED

-0.50

 vertrouwen

-0.50

djangoproject

-0.49

ErrUnexpectedEOF

-0.48

POSITIVE LOGITS

pt

1.66

pt

1.55

Pt

1.38

Pt

1.29

PT

1.22

PT

1.18

pts

0.92

pts

0.91

 platinum

0.76

Pts

0.72

Activations Density 0.214%

occurrences of specific formatting commands or notational symbols

No Comments

No Known Activations

occurrences of specific formatting commands or notational symbols

No Comments

No Known Activations