INDEX

Explanations

addresses and location-related terms

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

SequentialGroup

-0.54

le

-0.53

 open

-0.48

 staff

-0.48

Билгалдахарш

-0.48

 facility

-0.47

])),

-0.46

 Facility

-0.45

fchen

-0.44

cé

-0.44

POSITIVE LOGITS

Efq

0.88

 itſelf

0.86

 myſelf

0.81

 Chriftian

0.80

 fhort

0.76

 Jefus

0.76

Tazama

0.74

 ſeveral

0.73

 Houſe

0.72

 समीक्षक

0.72

Activations Density 0.379%

addresses and location-related terms

No Comments

No Known Activations