INDEX

Explanations

technical terms and concepts related to analysis and evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

ربای

-0.51

ciles

-0.49

-0.47

ashe

-0.46

COD

-0.45

crest

-0.44

انيف

-0.44

㎠

-0.43

Alike

-0.42

zenta

-0.42

POSITIVE LOGITS

 autorytatywna

0.68

Diwedd

0.67

aspectj

0.67

 présidenti

0.62

DebuggerStep

0.61

uxxxx

0.59

унд

0.59

BufferException

0.58

 препратки

0.58

 discorso

0.58

Activations Density 1.921%

technical terms and concepts related to analysis and evaluation

No Comments

No Known Activations

technical terms and concepts related to analysis and evaluation

No Comments

No Known Activations