INDEX

Explanations

specific coding-related terms and humor references

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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

gonic

-0.55

ByExample

-0.55

IDL

-0.53

NameInMap

-0.51

eof

-0.51

 Probe

-0.50

tomans

-0.50

ViewFeatures

-0.49

 BorderSide

-0.48

 push

-0.48

POSITIVE LOGITS

 multi

0.68

 Multi

0.68

multi

0.67

 laugh

0.63

 laughter

0.62

Multi

0.61

 chuckle

0.60

ๆ

0.60

 fauna

0.59

 multip

0.59

Activations Density 2.757%

specific coding-related terms and humor references

No Comments

No Known Activations