INDEX

Explanations

titles or positions within a hierarchical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 myſelf

-1.10

 houſe

-1.09

 itſelf

-1.07

 Theſe

-1.06

 pleaſure

-1.05

 ſche

-1.03

 Jefus

-0.99

 Houſe

-0.98

 greateſt

-0.96

 Majefty

-0.95

POSITIVE LOGITS

or

0.58

 last

0.55

<eos>

0.55

0.54

$\

0.53

0.51

0.50

in

0.48

Activations Density 0.471%

titles or positions within a hierarchical context

No Comments

No Known Activations