INDEX

Explanations

references to classification and categorization within various contexts

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

 referenties

-0.65

 <=",

-0.61

 ComVisible

-0.61

 springfox

-0.55

 saites

-0.53

 OFDb

-0.51

rawDesc

-0.50

lior

-0.48

toys

-0.48

 Twice

-0.48

POSITIVE LOGITS

 components

1.07

 categories

1.06

 main

1.06

 component

0.96

 phases

0.92

 major

0.92

 stages

0.89

 principaux

0.88

 layers

0.87

 Components

0.86

Activations Density 0.622%

references to classification and categorization within various contexts

No Comments

No Known Activations