INDEX

Explanations

patterns related to reasoning and justification

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

"[reason]" or "[reasons]" after certain words

np_acts-logits-general · gemini-2.0-flash

for any reason

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

SourceChecksum

-0.83

 estekak

-0.82

 Roskov

-0.73

 referenties

-0.72

 esternos

-0.65

 nahilalakip

-0.65

CppMethod

-0.63

Diwedd

-0.62

sätzlich

-0.61

########.

-0.59

POSITIVE LOGITS

 reason

3.02

 reasons

2.73

reason

2.41

Reason

2.17

 Reason

2.16

reasons

2.12

 Reasons

2.09

 REASON

2.02

Reasons

1.95

 REASONS

1.86

Activations Density 0.411%

patterns related to reasoning and justification

"[reason]" or "[reasons]" after certain words

for any reason

No Comments

No Known Activations