INDEX

Explanations

comparative and relational phrases, especially those indicating similarity or association

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

Comes before "like" or "unlike"

np_acts-logits-general · gemini-2.0-flash

like or unlike comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 pleaſure

-0.38

Batis

-0.37

OCCURRED

-0.37

 validamos

-0.37

pii

-0.35

 Avez

-0.34

väg

-0.33

 ecco

-0.33

 dise

-0.32

 Photocase

-0.32

POSITIVE LOGITS

unlike

0.78

Unlike

0.73

 unlike

0.72

Like

0.69

 Unlike

0.69

 Like

0.66

 LIKE

0.62

LIKE

0.61

like

0.60

endregion

0.59

Activations Density 0.280%

comparative and relational phrases, especially those indicating similarity or association

Comes before "like" or "unlike"

like or unlike comparisons

No Comments

No Known Activations