INDEX

Explanations

HTML heading tags and other structural markers in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

-0.57

her

-0.50

-0.49

AND

-0.48

Pe

-0.48

 Brock

-0.48

-0.47

’

-0.46

POSITIVE LOGITS

 ſche

1.15

 Houſe

1.10

 itſelf

1.10

 myſelf

1.07

 greateſt

1.03

 houſe

1.01

 Chriftian

1.00

 Anſ

1.00

 ſta

1.00

 pleaſure

1.00

Activations Density 0.141%

HTML heading tags and other structural markers in the text

No Comments

No Known Activations