INDEX

Explanations

occurrences of the word "no"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.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

↵↵

-0.61

-0.54

 Even

-0.53

<eos>

-0.53

-0.52

-0.51

↵

-0.51

zości

-0.50

All

-0.50

POSITIVE LOGITS

Efq

1.04

 enfans

0.91

 Jefus

0.90

 avoient

0.90

 étoient

0.86

 myſelf

0.85

IntoConstraints

0.81

 itſelf

0.80

bbene

0.80

UseVisualStyle

0.79

Activations Density 0.104%

occurrences of the word "no"

No Comments

No Known Activations