INDEX

Explanations

numerical values and their formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

webElementXpaths

-0.83

########.

-0.79

Diwedd

-0.76

SourceChecksum

-0.76

featureID

-0.75

 nonUne

-0.73

Rüyada

-0.73

rungsseite

-0.72

tagHelperRunner

-0.71

hoeddwyd

-0.70

POSITIVE LOGITS

enumi

0.67

0.59

[toxicity=0]

0.57

0.54

↵↵

0.54

setcounter

0.54

OrEmpty

0.54

‍♀️

0.53

etheless

0.53

↵

0.53

Activations Density 0.167%

numerical values and their formats

No Comments

No Known Activations