INDEX

Explanations

contextual references to specific geographical locations or structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 glisser

-0.58

ztat

-0.58

 balik

-0.56

 cliquant

-0.55

unsubscribe

-0.55

HNO

-0.55

-0.53

anyl

-0.51

ниципа

-0.51

Loose

-0.50

POSITIVE LOGITS

 itſelf

1.17

 myſelf

1.10

 raiſ

0.98

 Jefus

0.97

 becauſe

0.94

 faſt

0.92

 pleaſure

0.91

 themſelves

0.89

 ſche

0.88

 juſ

0.87

Activations Density 3.495%

contextual references to specific geographical locations or structures

No Comments

No Known Activations