INDEX

Explanations

references to various agencies and their actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

toPromise

-0.74

 tenis

-0.70

mort

-0.64

born

-0.64

scor

-0.63

-0.60

minWidth

-0.60

φ

-0.60

ness

-0.59

 Clim

-0.57

POSITIVE LOGITS

 agency

2.41

 agencies

2.40

 Agency

2.29

 AGENCY

2.21

 Agencies

2.18

agency

2.11

Agency

2.07

agencies

2.05

Agencies

1.96

 agencia

1.90

Activations Density 0.077%

references to various agencies and their actions

No Comments

No Known Activations