INDEX

Explanations

specific legal or procedural terminology and details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

def

-0.45

-0.44

 voud

-0.41

 home

-0.40

-0.39

ته

-0.38

New

-0.38

DIS

-0.37

occuper

-0.36

表示

-0.36

POSITIVE LOGITS

Personendaten

1.25

 myſelf

1.09

 purpoſe

1.09

featureID

0.99

 Majefty

0.96

 CreateTagHelper

0.95

 getItemId

0.95

 Monfieur

0.93

 pleaſure

0.92

 houſe

0.92

Activations Density 0.811%

specific legal or procedural terminology and details

No Comments

No Known Activations