INDEX

Explanations

statistical terms and numerical values in data or research contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

herjee

-0.48

rapeau

-0.43

 estekak

-0.43

SuccessListener

-0.42

jil

-0.42

 Secondary

-0.41

counters

-0.40

 counters

-0.40

ols

-0.40

ublik

-0.40

POSITIVE LOGITS

 Altogether

1.00

 resulting

1.00

resulting

0.99

 resultant

0.86

 Together

0.84

 Result

0.83

 thus

0.81

 ergibt

0.81

Together

0.80

thus

0.79

Activations Density 0.735%

statistical terms and numerical values in data or research contexts

No Comments

No Known Activations