INDEX

Explanations

expressions of previous experiences or knowledge that have not been encountered before

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 blijkt

-0.55

tidaknya

-0.55

 Inscrivez

-0.54

 quoque

-0.53

šech

-0.51

 câte

-0.51

 tambien

-0.50

 argint

-0.48

ânia

-0.48

stanford

-0.48

POSITIVE LOGITS

 ever

0.74

 jamais

0.69

 виправивши

0.69

 hitherto

0.67

 MainAxisSize

0.66

 Commencez

0.65

DoubleQuotes

0.64

providedIn

0.64

 BrowserModule

0.64

 незавершена

0.63

Activations Density 0.294%

expressions of previous experiences or knowledge that have not been encountered before

No Comments

No Known Activations