INDEX

Explanations

references to government operations and financial transactions

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

 affari

-0.39

skj

-0.35

faßt

-0.35

 dovol

-0.35

 hlad

-0.34

oredCriteria

-0.33

 expédi

-0.33

 desierto

-0.33

 détruire

-0.33

íslu

-0.33

POSITIVE LOGITS

 state

1.11

 estatal

1.09

 estatales

0.95

state

0.94

 State

0.88

State

0.82

 STATE

0.81

 staat

0.81

STATE

0.73

 государ

0.72

Activations Density 0.285%

references to government operations and financial transactions

No Comments

No Known Activations