INDEX

Explanations

key terms related to systems and frameworks in various contexts

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

BufferException

-0.76

 AssemblyTitle

-0.73

ſicht

-0.73

 يتيمه

-0.72

[@BOS@]

-0.68

<unused79>

-0.68

<unused41>

-0.68

<unused23>

-0.68

<unused3>

-0.68

<unused8>

-0.68

POSITIVE LOGITS

 demand

0.27

デンス

0.24

now

0.24

illerato

0.23

 côtés

0.23

 juntas

0.23

 menge

0.22

才是

0.22

 version

0.22

更

0.21

Activations Density 0.056%

key terms related to systems and frameworks in various contexts

No Comments

No Known Activations