INDEX

Explanations

numerical data and statistical results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

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

зулта

-0.63

rungsseite

-0.61

 بيها

-0.56

 مشين

-0.55

Bronnen

-0.53

 оригіналу

-0.52

 disambiguazione

-0.52

 doInBackground

-0.52

 خارجية

-0.51

 мәкал

-0.51

POSITIVE LOGITS

نر

0.59

 Normdatei

0.49

underland

0.47

#+#

0.47

 apoptotic

0.46

uala

0.45

 Unwin

0.45

pektive

0.44

ResponseWriter

0.44

hou

0.44

Activations Density 1.700%

numerical data and statistical results

No Comments

No Known Activations