INDEX

Explanations

terms associated with comebacks or recoveries in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 GetEnumerator

-0.82

Efq

-0.73

 oprot

-0.72

Obrázky

-0.71

 perpetuity

-0.65

 HttpNotFound

-0.65

menistan

-0.65

webElement

-0.65

 houſe

-0.63

 Houſe

-0.62

POSITIVE LOGITS

 recovery

0.67

 recover

0.58

 rescue

0.56

 Recovery

0.56

recovery

0.55

 surprises

0.55

 miracle

0.54

 late

0.53

 unexpected

0.53

 RECOVERY

0.53

Activations Density 0.461%

terms associated with comebacks or recoveries in various contexts

No Comments

No Known Activations