INDEX

Explanations

references to various models or examples in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

DispatchToProps

-0.74

Obf

-0.68

Чу

-0.66

achten

-0.63

mailto

-0.62

ähän

-0.58

 OnTrigger

-0.58

[&

-0.58

Eup

-0.57

 Kear

-0.57

POSITIVE LOGITS

 model

2.74

 MODEL

2.59

 Model

2.57

 models

2.55

model

2.36

 Models

2.32

MODEL

2.31

Model

2.20

 MODELS

2.13

models

2.06

Activations Density 0.092%

references to various models or examples in different contexts

No Comments

No Known Activations