INDEX

Explanations

segments where the number six is prominently mentioned

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

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

-0.68

qu

-0.67

-0.64

-0.59

der

-0.59

la

-0.58

et

-0.55

il

-0.54

le

-0.53

 last

-0.53

POSITIVE LOGITS

 itſelf

1.09

 Monfieur

1.08

 purpoſe

1.04

 Majefty

0.99

 myſelf

0.96

 Reſ

0.93

 ―――――

0.90

 bershka

0.88

 Jefus

0.87

 Houſe

0.86

Activations Density 0.253%

segments where the number six is prominently mentioned

No Comments

No Known Activations