INDEX

Explanations

references to minimum thresholds or limits in various contexts

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

 AssemblyCompany

-0.86

 مرئيه

-0.66

[]:

-0.66

 otomatig

-0.62

IndentedString

-0.61

protoimpl

-0.57

[@"

-0.56

 surla

-0.56

åt

-0.55

 IndexPath

-0.55

POSITIVE LOGITS

 achievable

0.84

 amount

0.80

 possible

0.78

 value

0.74

值

0.73

 allowable

0.71

 attainable

0.71

 POSSIBLE

0.69

 poffible

0.69

 leeftijd

0.68

Activations Density 0.179%

references to minimum thresholds or limits in various contexts

No Comments

No Known Activations