INDEX

Explanations

terms related to disclosure and stairs

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

ly

-0.59

を発

-0.57

ůvod

-0.55

Dis

-0.54

tvrt

-0.53

tation

-0.53

Wür

-0.52

-0.51

hyd

-0.51

protoc

-0.51

POSITIVE LOGITS

 stairs

0.92

stairs

0.88

 resize

0.85

 Resize

0.84

 Stairs

0.80

resize

0.79

Personendaten

0.78

Yag

0.77

ExecuteReader

0.77

 FormsModule

0.76

Activations Density 0.093%

terms related to disclosure and stairs

No Comments

No Known Activations