INDEX

Explanations

mentions of political figures and their actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 obviously

-0.95

 reportedly

-0.93

obviously

-0.90

 seem

-0.86

 Obviously

-0.83

 seems

-0.82

 Seems

-0.82

 said

-0.82

Seems

-0.82

seems

-0.81

POSITIVE LOGITS

 indeed

0.87

 inderdaad

0.75

indeed

0.71

 وأن

0.68

 efectivamente

0.63

 actually

0.60

Indeed

0.57

 Indeed

0.56

 tatsächlich

0.56

Actually

0.55

Activations Density 1.620%

mentions of political figures and their actions

No Comments

No Known Activations