INDEX

Explanations

repeated instances of the letter 'e'

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

}|\

-0.56

 ISNI

-0.55

 triom

-0.55

 transparence

-0.53

 Hammond

-0.53

,:]

-0.53

ъм

-0.52

muş

-0.52

 excer

-0.51

}^\

-0.51

POSITIVE LOGITS

2.62

1.12

ei

1.00

ete

0.96

eeeee

0.90

eing

0.87

 resourceCulture

0.87

️

0.86

ejo

0.86

eet

0.86

Activations Density 0.147%

repeated instances of the letter 'e'

No Comments

No Known Activations