INDEX

Explanations

concepts related to beliefs and their validity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 оригіналу

-0.61

ThroughAttribute

-0.57

tablir

-0.52

 IndexError

-0.52

EndContext

-0.52

HideInInspector

-0.52

montre

-0.51

GeoNames

-0.51

 reinvent

-0.49

 notations

-0.49

POSITIVE LOGITS

liciter

0.68

 ought

0.65

 claims

0.57

 claim

0.56

 account

0.55

strual

0.55

НОСТИ

0.55

arela

0.55

edly

0.54

 claimed

0.54

Activations Density 0.473%

concepts related to beliefs and their validity

No Comments

No Known Activations