INDEX

Explanations

connectives that indicate additional information or similarity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/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.34.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

 للاسماء

-0.62

 nakalista

-0.57

RuleContext

-0.56

 Lordships

-0.55

DataAnnotations

-0.55

 ComVisible

-0.53

MockMvc

-0.52

 autorytatywna

-0.50

Хьажоргаш

-0.49

GEBURTSDATUM

-0.49

POSITIVE LOGITS

nd

0.87

বং

0.65

nd

0.63

ND

0.54

Nd

0.49

and

0.48

AND

0.46

Nd

0.44

0.43

ND

0.43

Activations Density 0.157%

connectives that indicate additional information or similarity

No Comments

No Known Activations