INDEX

Explanations

terms related to barriers and limitations 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 @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 acquittal

-0.61

 gynhyrchwyd

-0.60

 twij

-0.56

 hasard

-0.54

 vapa

-0.54

jenih

-0.53

localtime

-0.53

EDEFAULT

-0.52

 friable

-0.52

 ouvertes

-0.52

POSITIVE LOGITS

 blocking

1.46

 restricting

1.45

 blocked

1.38

 restrict

1.38

 shut

1.28

 restrictive

1.28

 restricts

1.28

 closed

1.26

 shutting

1.25

 restricted

1.25

Activations Density 0.902%

terms related to barriers and limitations in various contexts

No Comments

No Known Activations