INDEX

Explanations

numerical values and parameters within a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 Moser

-0.59

 fermés

-0.55

 محفوظة

-0.53

 Autorizaciones

-0.53

 Mohr

-0.53

clerView

-0.53

 passés

-0.52

 meurt

-0.52

 rendus

-0.51

 Cæsar

-0.51

POSITIVE LOGITS

LookAnd

0.60

 doInBackground

0.59

 IndexPath

0.55

RuleContext

0.49

ftagPool

0.48

 belf

0.47

setof

0.47

instancetype

0.46

 isempty

0.46

estad

0.45

Activations Density 0.659%

numerical values and parameters within a technical context

No Comments

No Known Activations