INDEX

Explanations

expressions related to identity and relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Jereo

-0.51

########.

-0.44

 TAMBIÉN

-0.41

 saddest

-0.41

 intolerable

-0.41

 unfavourable

-0.41

 shenanigans

-0.38

 prohibitive

-0.38

 antics

-0.37

 offending

-0.37

POSITIVE LOGITS

 sleek

0.84

 crisp

0.80

 bright

0.79

 clean

0.76

 calm

0.72

 fast

0.72

 vibrant

0.71

 warm

0.69

明るく

0.69

 energetic

0.67

Activations Density 0.482%

expressions related to identity and relationships

No Comments

No Known Activations