INDEX

Explanations

references to legal terminology and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

-0.46

 opening

-0.45

ficie

-0.42

 withdrawing

-0.42

DONE

-0.41

PerformLayout

-0.40

 increasing

-0.40

人了

-0.40

 Opening

-0.39

்க்க

-0.39

POSITIVE LOGITS

oneofs

0.80

CloseOperation

0.80

WriteBarrier

0.80

Datuak

0.80

__':

0.76

 will

0.74

is

0.74

GeneratedMessage

0.74

حياته

0.73

 would

0.73

Activations Density 3.781%

references to legal terminology and processes

No Comments

No Known Activations