INDEX

Explanations

words related to favorability or preference

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ftet

-0.83

nought

-0.68

足

-0.64

waypoint

-0.62

Giao

-0.60

cos

-0.59

cop

-0.59

Abitanti

-0.59

 HEND

-0.59

scrolling

-0.58

POSITIVE LOGITS

 favours

1.11

 favour

1.09

 neighbourhoods

1.05

 sceptre

0.95

 Favor

0.95

 fibres

0.95

 honours

0.95

Favor

0.95

 honoured

0.94

 disambiguazione

0.94

Activations Density 0.278%

words related to favorability or preference

No Comments

No Known Activations