INDEX

Explanations

terms related to accessibility

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

ease of access

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

CppMethod

-0.33

astify

-0.31

 setCode

-0.31

magitan

-0.31

 didReceive

-0.31

wirtschaft

-0.31

FormState

-0.31

 algoritmo

-0.31

 potentially

-0.30

dollis

-0.30

POSITIVE LOGITS

httphttps

0.64

 propOrder

0.64

 Easily

0.63

accessible

0.62

 Accessible

0.62

Easily

0.61

uxxxx

0.61

 access

0.59

 accessible

0.59

Accessible

0.57

Activations Density 0.006%

terms related to accessibility

ease of access

No Comments

No Known Activations