INDEX

Explanations

references to political roles and positions within organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

 lenker

-0.75

tagHelperRunner

-0.72

дравствуйте

-0.69

 GeoNames

-0.68

 gambe

-0.68

InvalidProtocol

-0.67

protoc

-0.65

sizeCache

-0.65

httphttps

-0.64

 Bourgoin

-0.64

POSITIVE LOGITS

job

0.88

 jobs

0.68

job

0.65

Job

0.65

 positions

0.63

JOB

0.63

 position

0.62

 hold

0.62

 Hold

0.62

Job

0.61

Activations Density 0.268%

references to political roles and positions within organizations

No Comments

No Known Activations