INDEX

Explanations

descriptions of methodologies or strategies

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

AutoField

-0.54

tys

-0.52

My

-0.51

 presence

-0.50

 Super

-0.49

LETED

-0.49

MetaType

-0.49

 Blackman

-0.48

rotum

-0.48

POSITIVE LOGITS

 approaches

1.13

 Approaches

1.07

 Approach

0.97

 APPROACH

0.97

 approach

0.93

approach

0.86

 pinulongan

0.86

Approach

0.86

Appro

0.78

 Monfieur

0.77

Activations Density 0.084%

descriptions of methodologies or strategies

No Comments

No Known Activations