INDEX

Explanations

references to financial results and performance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

styleType

-0.70

parsedMessage

-0.70

IUrlHelper

-0.69

 betweenstory

-0.63

DllImport

-0.62

ISupport

-0.62

Aholisi

-0.62

httphttps

-0.61

 propOrder

-0.60

Datuak

-0.60

POSITIVE LOGITS

 year

0.57

 October

0.57

 July

0.57

 during

0.56

 January

0.54

 December

0.51

 September

0.51

 November

0.49

 throughout

0.47

 August

0.47

Activations Density 0.044%

references to financial results and performance

No Comments

No Known Activations