INDEX

Explanations

references to average values in data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

ulemon

-0.65

 DialogInterface

-0.62

sp

-0.60

 Murdoch

-0.56

ed

-0.56

hasMoreElements

-0.55

sk

-0.55

emb

-0.54

 Kirk

-0.54

führt

-0.54

POSITIVE LOGITS

 average

3.23

average

3.04

 Average

3.03

Average

2.96

 AVERAGE

2.85

 averages

2.64

AVERAGE

2.62

 averaged

2.55

avg

2.46

 averaging

2.46

Activations Density 0.078%

references to average values in data

No Comments

No Known Activations