INDEX

Explanations

start tokens and sections of text indicating new content or topics

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

preceding "from" or "to"

np_acts-logits-general · gemini-2.0-flash

from various sources

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

Datuak

-0.78

Numerology

-0.69

$.

-0.63

"):

-0.61

 pleaſure

-0.61

Rè

-0.61

 Monfieur

-0.60

__*/

-0.59

 myſelf

-0.59

POSITIVE LOGITS

 these

0.74

 właśnie

0.67

 právě

0.67

 this

0.60

these

0.60

 именно

0.59

 этих

0.58

 THESE

0.56

aarrggbb

0.51

 diesen

0.51

Activations Density 0.458%

start tokens and sections of text indicating new content or topics

preceding "from" or "to"

from various sources

No Comments

No Known Activations