INDEX

Explanations

patterns and metrics from experimental results or data analysis

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

scientific abbreviations and other languages

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.32.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 juſt

-0.53

Chham

-0.52

 juft

-0.51

 perſon

-0.50

 ſta

-0.50

 ſte

-0.49

CPtr

-0.47

 myſelf

-0.47

 fubject

-0.47

 canon

-0.45

POSITIVE LOGITS

:✨

0.50

 مشين

0.48

seamnă

0.42

oarece

0.41

 infatti

0.38

 entanto

0.37

 AssemblyCulture

0.37

 adolescente

0.36

RTEE

0.36

 BoxDecoration

0.35

Activations Density 0.031%

patterns and metrics from experimental results or data analysis

scientific abbreviations and other languages

No Comments

No Known Activations