INDEX

Explanations

phrases and terms related to comparative analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 Majefty

-0.66

 houſe

-0.63

 pleaſure

-0.60

 itſelf

-0.58

 enfans

-0.57

 beſt

-0.55

 ſche

-0.55

ſelf

-0.55

 fubject

-0.55

 Houſe

-0.54

POSITIVE LOGITS

 UPON

0.71

the

0.68

against

0.65

 upon

0.63

Against

0.63

Diwedd

0.61

upon

0.61

 various

0.60

 Against

0.59

 forState

0.57

Activations Density 0.513%

phrases and terms related to comparative analysis

No Comments

No Known Activations