INDEX

Explanations

viewer ratings and audience metrics for television shows

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

ratings uptick

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

BagLayout

-0.49

WithMany

-0.45

NameValuePair

-0.40

glement

-0.39

iltà

-0.37

mangle

-0.35

Geplaatst

-0.35

 pulumi

-0.34

 Herzens

-0.34

optString

-0.33

POSITIVE LOGITS

 televisiva

0.59

 viewers

0.58

television

0.56

 television

0.55

 ratings

0.55

 telewiz

0.52

 Television

0.51

andExpect

0.51

tagext

0.51

 télévis

0.50

Activations Density 0.016%

viewer ratings and audience metrics for television shows

ratings uptick

No Comments

No Known Activations