INDEX

Explanations

references to cleanliness or clean-up activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 esternos

-0.70

abestanden

-0.62

ServletException

-0.61

__":

-0.61

bootstrapcdn

-0.60

usercontent

-0.59

 lác

-0.58

__':

-0.58

 AssemblyCulture

-0.57

 spéciales

-0.55

POSITIVE LOGITS

liness

0.99

 slate

0.82

 sweep

0.76

 Auge

0.69

 Sweep

0.63

ups

0.61

Sweep

0.61

 Slate

0.58

 tidy

0.58

room

0.58

Activations Density 0.095%

references to cleanliness or clean-up activities

No Comments

No Known Activations