INDEX

Explanations

code syntax and structural components related to programming

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

pages and specific items

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

httphttps

-0.85

UrlResolution

-0.82

 '\\;'

-0.81

 EconPapers

-0.79

ValueStyle

-0.76

findpost

-0.72

 propOrder

-0.72

MessageTagHelper

-0.71

GraphicsUnit

-0.69

jsonwebtoken

-0.69

POSITIVE LOGITS

kui

0.28

ękuję

0.28

GeoNames

0.27

 both

0.27

]=>

0.27

0.26

↵

0.25

Activations Density 0.166%

code syntax and structural components related to programming

pages and specific items

No Comments

No Known Activations