INDEX

Explanations

phrases expressing reasons or justifications

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

Following "for"

np_acts-logits-general · gemini-2.0-flash

for followed by pronoun or gerund

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 متعلقه

-0.73

 gainera

-0.59

 للمعارف

-0.57

WithIOException

-0.56

Getenv

-0.53

bkz

-0.53

Jegyzetek

-0.51

āci

-0.51

InsertCommand

-0.50

emptory

-0.49

POSITIVE LOGITS

 being

0.99

 having

0.85

being

0.80

having

0.79

 étant

0.76

 haberse

0.75

avoir

0.73

 esser

0.72

 habiendo

0.72

Being

0.71

Activations Density 0.191%

phrases expressing reasons or justifications

Following "for"

for followed by pronoun or gerund

No Comments

No Known Activations