INDEX

Explanations

phrases that indicate repetition or redundancy in various contexts

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

Comes before a negative or undesirable outcome

np_acts-logits-general · gemini-2.0-flash

another followed by a descriptor

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

ponses

-0.60

 lenker

-0.55

featureID

-0.54

onViewCreated

-0.54

Filmografia

-0.52

RetentionPolicy

-0.52

 AssemblyTitle

-0.51

artis

-0.51

ГЛА

-0.50

 Spade

-0.50

POSITIVE LOGITS

 another

0.85

 очеред

0.85

 Another

0.80

 again

0.77

another

0.75

Another

0.75

 lagi

0.71

 ANOTHER

0.70

 Again

0.69

又是

0.68

Activations Density 0.164%

phrases that indicate repetition or redundancy in various contexts

Comes before a negative or undesirable outcome

another followed by a descriptor

No Comments

No Known Activations