INDEX

Explanations

occurrences of the word "saw" in various forms, indicating a focus on visual perception or the act of seeing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Grunde

-0.58

 <<<<<<<<<<<<<<

-0.57

eol

-0.57

pati

-0.56

 dinh

-0.54

oglobulin

-0.53

LOAT

-0.53

 municipales

-0.52

 thân

-0.51

estin

-0.51

POSITIVE LOGITS

saw

3.76

saw

3.64

Saw

3.46

Saw

3.34

SAW

2.90

 saws

2.34

SAW

2.21

 sawing

1.75

 såg

1.59

 Sawyer

1.59

Activations Density 0.083%

occurrences of the word "saw" in various forms, indicating a focus on visual perception or the act of seeing

No Comments

No Known Activations