INDEX

Explanations

discussions related to updates or changes in experimental conditions and their effects

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

Word after "predominance"

np_acts-logits-general · gemini-2.0-flash

predominance and concentration

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/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.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 INTEGR

-0.50

setY

-0.49

[--

-0.48

izeit

-0.47

ump

-0.46

Autre

-0.46

 Unto

-0.46

 permanently

-0.45

 tế

-0.45

permanent

-0.45

POSITIVE LOGITS

 disproportion

0.81

 concentrate

0.79

 preferentially

0.78

 concentrated

0.76

 concentrating

0.76

 konzent

0.74

 estekak

0.74

concentr

0.73

PerformLayout

0.73

 koncentr

0.71

Activations Density 0.705%

discussions related to updates or changes in experimental conditions and their effects

Word after "predominance"

predominance and concentration

No Comments

No Known Activations