INDEX

Explanations

instances of the word "large" in varying contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 AssemblyTitle

-0.88

ecap

-0.81

IElement

-0.73

 obligé

-0.69

ècie

-0.68

StructEnd

-0.68

FTFY

-0.68

 auxquelles

-0.68

 ISSUED

-0.67

 complémentaires

-0.67

POSITIVE LOGITS

LARGE

0.88

 Large

0.87

 enough

0.84

Large

0.82

 LARGE

0.81

Big

0.75

 Marge

0.73

 large

0.73

 Small

0.72

Big

0.72

Activations Density 0.090%

instances of the word "large" in varying contexts

No Comments

No Known Activations