INDEX

Explanations

programming concepts related to data handling and manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 houſe

-0.79

 pleaſure

-0.72

ſelf

-0.67

 ſche

-0.67

 Houſe

-0.66

 ſtate

-0.62

 perſon

-0.58

 ſtre

-0.58

 ſch

-0.57

 purpoſe

-0.57

POSITIVE LOGITS

ized

0.76

aced

0.73

ched

0.73

ded

0.72

ted

0.71

tered

0.69

cled

0.69

ourced

0.69

ked

0.68

tened

0.68

Activations Density 2.868%

programming concepts related to data handling and manipulation

No Comments

No Known Activations