INDEX

Explanations

elements related to technical error messages or warnings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Personendaten

-0.79

 <<<<<<<<<<<<<<

-0.65

 Diſ

-0.59

 hObject

-0.58

 purpoſe

-0.57

 ſmall

-0.57

 houſe

-0.55

 fhort

-0.55

 fubject

-0.55

 dout

-0.55

POSITIVE LOGITS

Legende

0.60

Teb

0.60

Se

0.60

reo

0.59

 Keene

0.58

Te

0.58

 WebElement

0.57

dega

0.57

 LEED

0.57

reza

0.57

Activations Density 0.856%

elements related to technical error messages or warnings

No Comments

No Known Activations