INDEX

Explanations

references to input prompts or actions involving entering data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

дмила

-0.57

PyExc

-0.56

fromCharCode

-0.52

 cetera

-0.50

جوايز

-0.49

lhs

-0.48

habad

-0.48

itus

-0.48

ândia

-0.47

lujah

-0.47

POSITIVE LOGITS

PRO

0.91

Enter

0.87

ro

0.87

pro

0.77

enter

0.75

 enter

0.74

 nahilalakip

0.74

RO

0.70

 adoptive

0.70

ro

0.69

Activations Density 0.108%

references to input prompts or actions involving entering data

No Comments

No Known Activations