INDEX

Explanations

information related to software development issues and troubleshooting

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

question asks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 queſta

-1.18

 myſelf

-1.16

 itſelf

-1.06

ſelves

-1.05

 ſever

-1.04

ſy

-1.01

 ſte

-1.00

 ſeveral

-0.99

 eſſ

-0.98

 faſt

-0.97

POSITIVE LOGITS

 alot

0.64

 thier

0.59

0.58

 diatas

0.57

 loosing

0.57

 thats

0.55

 dimana

0.54

 didalam

0.54

 didnt

0.53

Activations Density 7.352%

information related to software development issues and troubleshooting

question asks

No Comments

No Known Activations

information related to software development issues and troubleshooting

question asks

No Comments

No Known Activations