INDEX

Explanations

phrases related to functionality and operability 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

<bos>

-0.61

 namorados

-0.60

 regalías

-0.54

rollment

-0.54

 bendera

-0.53

ItemBackground

-0.53

nalpot

-0.53

 mídia

-0.52

 profundidad

-0.52

 enggak

-0.52

POSITIVE LOGITS

Working

1.09

 working

1.07

working

1.06

 Working

1.03

WORKING

0.93

 WORKING

0.90

 worker

0.74

worker

0.71

Worker

0.69

 Worker

0.69

Activations Density 0.013%

phrases related to functionality and operability in various contexts

No Comments

No Known Activations