INDEX

Explanations

technical terms and phrases related to programming errors and file operations

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

OGND

-1.09

richTextPanel

-0.90

 CreateTagHelper

-0.90

 AssemblyTitle

-0.85

TagMode

-0.82

 estekak

-0.82

 ProtoMessage

-0.81

WebVitals

-0.77

⟬

-0.74

GraphicsUnit

-0.73

POSITIVE LOGITS

↵

0.35

Mal

0.35

↵↵

0.33

CA

0.32

 thing

0.31

dit

0.31

here

0.30

ine

0.30

 Western

0.30

Activations Density 0.035%

technical terms and phrases related to programming errors and file operations

No Comments

No Known Activations