INDEX

Explanations

instances of the word "utilization."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

__(/*!

-0.93

 المعيارى

-0.76

 ExecuteAsync

-0.75

TabStop

-0.65

SpringRunner

-0.64

 виправивши

-0.62

ReusableCell

-0.62

 NSCoder

-0.61

StoryboardSegue

-0.60

Джерела

-0.60

POSITIVE LOGITS

 żel

0.64

 Guinée

0.62

bacillus

0.59

Aholisi

0.58

risch

0.57

وصلات

0.56

Geplaatst

0.56

heets

0.54

ilan

0.53

тиво

0.53

Activations Density 0.002%

instances of the word "utilization."

No Comments

No Known Activations