INDEX

Explanations

instances of criticism or negative evaluations of entities or actions

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

followed by citations or URLs

np_acts-logits-general · gemini-2.0-flash

words expressing criticism or debate

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

maakt

-0.32

 artig

-0.27

errHandler

-0.27

кновен

-0.27

win

-0.27

發生

-0.27

OGND

-0.27

Ums

-0.26

体现

-0.26

 oprot

-0.26

POSITIVE LOGITS

 nakalista

0.73

VersionUID

0.69

SequentialGroup

0.67

givings

0.66

figsize

0.63



0.62



0.61

 onPostExecute

0.60

 estekak

0.59

VolleyError

0.59

Activations Density 0.693%

instances of criticism or negative evaluations of entities or actions

followed by citations or URLs

words expressing criticism or debate

No Comments

No Known Activations