INDEX

Explanations

negative descriptors related to failures or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

setVerticalGroup

-0.72

WireFormatLite

-0.67

DoubleQuotes

-0.63

 ArrayAdapter

-0.61

 utafitiHapana

-0.60

 getragen

-0.59

Portail

-0.56

isNotBlank

-0.56

 uppermost

-0.56

 reciprocity

-0.56

POSITIVE LOGITS

 dangerous

0.81

 negativ

0.68

 invalid

0.67

 threatening

0.66

 toxic

0.66

 dangerously

0.64

 worse

0.64

Worse

0.64

 Worse

0.61

 harmful

0.61

Activations Density 1.715%

negative descriptors related to failures or issues

No Comments

No Known Activations