INDEX

Explanations

mathematical notation and specific formatting typically used in equations and algorithms

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

 pleaſure

-1.25

 Monfieur

-1.16

 raiſ

-1.15

 purpoſe

-1.14

 houſe

-1.13

 ſtate

-1.11

 Anſ

-1.10

 ſever

-1.09

 Jefus

-1.08

Efq

-1.07

POSITIVE LOGITS

0.63

0.62

0.57

">(</

0.56

0.55

’

0.54

0.48

ברס

0.47

люби

0.47

Activations Density 0.401%

mathematical notation and specific formatting typically used in equations and algorithms

No Comments

No Known Activations