INDEX

Explanations

unique identifiers or codes associated with data entries

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

Hexadecimal characters and numbers

np_acts-logits-general · gemini-2.0-flash

ids and codes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

Демографія

-0.54

 imageNamed

-0.52

горо

-0.50

ocry

-0.49

 <=",

-0.49

atient

-0.49

rettes

-0.48

epres

-0.48

 Persson

-0.48

orrho

-0.47

POSITIVE LOGITS

 Roskov

0.85

 opérateur

0.69

 calendriers

0.68

derung

0.62

Geplaatst

0.58

OGND

0.58

 hâte

0.55

">//

0.55

nô

0.55

 jsPsych

0.55

Activations Density 0.084%

unique identifiers or codes associated with data entries

Hexadecimal characters and numbers

ids and codes

No Comments

No Known Activations