INDEX

Explanations

key terms related to conflicts and their resolutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 raiſ

-0.98

Efq

-0.88

 poffe

-0.87

 whoſe

-0.86

toThrow

-0.86

 Monfieur

-0.84

 myſelf

-0.83

 difp

-0.82

 Majefty

-0.82

 uſed

-0.81

POSITIVE LOGITS

 unfold

0.51

 unfolds

0.50

 arise

0.48

 materialize

0.48

 occur

0.47

 happen

0.46

 estimés

0.45

0.44

 happens

0.42

 come

0.42

Activations Density 0.480%

key terms related to conflicts and their resolutions

No Comments

No Known Activations