INDEX

Explanations

negative statements or expressions of doubt

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/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.33.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

 soient

-0.77

 puissent

-0.65

 aient

-0.63

 siano

-0.62

Be

-0.61

Be

-0.59

 betweenstory

-0.57

RegressionTest

-0.53

 бъдат

-0.53

sidemargin

-0.53

POSITIVE LOGITS

 belong

0.88

 need

0.87

 want

0.81

 seem

0.81

 deserve

0.80

 exist

0.75

 know

0.72

owe

0.66

 mean

0.62

 require

0.59

Activations Density 0.206%

negative statements or expressions of doubt

No Comments

No Known Activations