INDEX

Explanations

structured exploratory phrases and references to studies or group comparisons

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

quantities and enumerations

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IBOutlet

-0.49

UnusedPrivate

-0.46

Personendaten

-0.46

 EconPapers

-0.46

SequentialGroup

-0.45

 רבים

-0.44

enumi

-0.43

 bezeichneter

-0.41

providedIn

-0.40

 Shallow

-0.38

POSITIVE LOGITS

ArrowToggle

0.82

four

0.73

 four

0.73

five

0.71

 five

0.68

 CreateTagHelper

0.63

Five

0.60

 quatre

0.60

 quatro

0.60

six

0.59

Activations Density 0.412%

structured exploratory phrases and references to studies or group comparisons

quantities and enumerations

No Comments

No Known Activations