INDEX

Explanations

references to websites and online resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

-0.55

hnia

-0.55

-0.53

↵↵

-0.52

 form

-0.52

-0.51

GRANTED

-0.49

<eos>

-0.47

-0.43

POSITIVE LOGITS

ſelf

1.15

 myſelf

1.01

 Beſ

1.01

Efq

0.98

 itſelf

0.95

 Reſ

0.95

 ſmall

0.94

/**

0.93

ſelves

0.93

 Jefus

0.93

Activations Density 0.279%

references to websites and online resources

No Comments

No Known Activations