INDEX

Explanations

references to the second-person perspective

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

ImageContext

-0.45

ERVIEW

-0.43

WebElementEntity

-0.42

MigrationBuilder

-0.41

sweise

-0.40

Pautan

-0.39

úgó

-0.37

AutoField

-0.37

setVerticalGroup

-0.36

 resultó

-0.35

POSITIVE LOGITS

machine

0.50

 machine

0.46

machines

0.45

 chatbot

0.42

ſchen

0.41

retra

0.41

 carousel

0.40

 machines

0.40

paka

0.40

 declar

0.40

Activations Density 0.041%

references to the second-person perspective

No Comments

No Known Activations