INDEX

Explanations

phrases related to justifications or explanations for actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

contentLoaded

-0.63

emptory

-0.60

bkz

-0.56

everyone

-0.55

ranslated

-0.54

 kvar

-0.54

WithIOException

-0.53

piram

-0.53

falgar

-0.53

TintMode

-0.53

POSITIVE LOGITS

 having

1.10

 being

1.06

having

0.94

 haberse

0.86

Having

0.84

 étant

0.82

 esser

0.81

being

0.79

 Having

0.79

 essere

0.77

Activations Density 0.193%

phrases related to justifications or explanations for actions

No Comments

No Known Activations