INDEX

Explanations

statements emphasizing the concept of "fact" and its relevance in a discussion

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

"fact"

np_acts-logits-general · gemini-2.0-flash

fact followed by 'is' or 'that'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 kuni

-0.69

 AssemblyCompany

-0.69

ρους

-0.69

ised

-0.67

Autoritní

-0.65

 समीक्षक

-0.65

ρίς

-0.63

gyz

-0.63

ագրություններ

-0.61

 initComponents

-0.61

POSITIVE LOGITS

 fact

0.94

 FACT

0.79

fact

0.79

 Fact

0.75

FACT

0.72

 facts

0.72

 Facts

0.71

 FACTS

0.63

Fact

0.63

facts

0.62

Activations Density 0.097%

statements emphasizing the concept of "fact" and its relevance in a discussion

"fact"

fact followed by 'is' or 'that'

No Comments

No Known Activations