INDEX

Explanations

negative statements and contradictions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 jsPsych

-0.84

 Roskov

-0.69

abetes

-0.61

erialization

-0.58

 OFDb

-0.54

titleMargin

-0.54

 surla

-0.52

transQ

-0.52

chkin

-0.51

uramente

-0.49

POSITIVE LOGITS

 other

0.60

 remaining

0.55

 much

0.55

 remain

0.53

findpost

0.53

 remains

0.51

 никакого

0.49

ellschaft

0.49

колко

0.48

人には

0.47

Activations Density 0.488%

negative statements and contradictions

No Comments

No Known Activations